Toshiba T3100/20 with a dead drive?

[ragingbaboon38]

I've recently picked up a Toshiba T3100/20 from a seller on eBay, who said that they found the computer in an estate sale and they didn't know much about it. They were unable to test the system due to the lack of a system disk (which, in the images on the eBay post, the machine was displaying a typical "Insert system disk and press any key" message.) I figured this would be an easy fix, just write a disk that's bootable.

Thankfully, it arrived rather quickly and it's in great shape, no cosmetic damage at all... although it smells weird. When I first started it up, the CMOS battery was dead, which is no surprise... although what did surprise me was that after dismissing the CMOS error messages, the machine stays on a blinking cursor for a minute or two before beeping and displaying a message reading "ERROR ENCOUNTERED INITIALIZING HARD DRIVE"... this stays for a few moments before disappearing and attempting to boot from floppy. Thankfully, that part works. I was able to run the diagnostic disk and check the CMOS settings... sadly, no matter what hard drive configurations I tried, I keep getting this hard drive error on boot. And it goes without saying that the drive can't be interacted with at all, the computer acts like there's nothing connected. I've tried listening for the drive, but the computer is so loud that I can't tell hard drive noise apart from all the other noise it makes, so I have no idea if the drive is even powering on. Strangely enough, using the "auto" config option on the setup program lists both the floppy drive and hard drive correctly. So I don't know if it's loading those settings from somewhere in ROM or if it's actually correctly detecting the hard drive. Either way, it's not working. Thankfully everything else works fine; the display looks beautiful, the keyboard feels nice, and the floppy disk drive still works fine for things like Planet X3.

I asked the seller about it too. There was no mention of this error message on the eBay page, nor were there pictures of it. Considering that the message appears before the "Insert system disk" message, and the computer very obnoxiously beeps for a solid minute to alert the user of said hard drive failure, I assumed that the seller would have at least mentioned it, but that might be too far of a reach. They said they never saw the message about the hard drive, so I have no idea why it's only appearing now. Maybe it was damaged during shipping, I've heard things about older hard drives needing certain care before being shipped.

What I'm wondering is if there's any possible way to diagnose this or fix this, preferably without taking the computer apart. I'll open it if I have to, I just really don't like opening laptops and I'm worried I'll break something.

 

[granzeier]

I used to use a tool called: idediag.exe (https://www.pcorner.com/list/HDUTIL/IDEDIAG.ZIP/INFO/). This would query the actual controller on an IDE drive, and report what it said. That helped me many times. If your drive is really working, this may get the info that you need from the drive itself.

There are, apparently, several different programs with that, or similar, names; but the one above is the one that I used (or at least it looks the same.)

 

[Brian Todd]

What I'm wondering is if there's any possible way to diagnose this or fix this, preferably without taking the computer apart. I'll open it if I have to, I just really don't like opening laptops and I'm worried I'll break something.

Hopefully idediag will help you out, but if the HDD is toast (which is very possible) then be very careful in taking the laptop apart as the plastic around the screw 'nuts' is extremely brittle indeed. Even being careful, I ended up using a lot of epoxy on my T3100 Having said that, the lubrication on the HDD has likely dried out as the unit has not been used for a long time (your post suggests this is the case), so it might be worth trying multiple short boots (power on and off) and this might just free up the platter enough for it to spin up .. worth a try at least. Let us know how you get on!

 

[modem7]

If I go to my car and start the engine, that does not mean that my car is fully functional. Similarly, if a program that asks the IDE drive for information returns information, that does not mean that the drive is fully functional.

 

[granzeier]

If I go to my car and start the engine, that does not mean that my car is fully functional. Similarly, if a program that asks the IDE drive for information returns information, that does not mean that the drive is fully functional.

That is true - I was not giving a complete fix to ragingbaboon38's troubles. My suggestion was just to help with possibly getting a bit more information from the drive itself, rather than possibly good/possibly bad info from the BIOS. I just wish that I could give him enough help to get his "car" driving down the highway.

 

[ragingbaboon38]

Thanks for the help, everyone.

Hopefully idediag will help you out, but if the HDD is toast (which is very possible) then be very careful in taking the laptop apart as the plastic around the screw 'nuts' is extremely brittle indeed. Even being careful, I ended up using a lot of epoxy on my T3100 Having said that, the lubrication on the HDD has likely dried out as the unit has not been used for a long time (your post suggests this is the case), so it might be worth trying multiple short boots (power on and off) and this might just free up the platter enough for it to spin up .. worth a try at least. Let us know how you get on!

I tried your "turn it off and on" idea and unfortunately that didn't get anything working. As for the other suggestions such as idediag, does it even work on non-IDE drives? I'm almost certain that the T3100 has a pre-IDE (MFM?) drive. Apologies if that makes me sound stupid cause I don't know the most about old storage formats.

I would have tested idediag already if it weren't for one thing... both of my machines capable of writing floppy disks aren't working! One is a Compaq Presario 1200, and the other some kind of HP desktop with Windows 2000 on it. I haven't had to use either machine in a while, and, as my luck goes, the Presario laptop won't turn on and the HP's floppy drive isn't working. So I'll have to get one of those two machines working and writing disks before I can get back to trying to solve this problem.

 

[Conventional Memories]

The JVC MFM HDD in these are very unreliable.
On the 10MB drives (in the T3100) I've been successful at repairing them by replacing the leaking caps on the bottom of the drive. And by low level formatting them with speedstor (this has helped me even when the Toshiba Test3 utility couldn't preform a LLformat).
On the 20MB drives (like in the T3100/20) I haven't found a way to get them to work reliably unfortunately.

Also sometimes these drives don't even spin up, or don't get up to speed and speed down. That can usually be fixed by dropping a bit of bearing oil into the spindle motor and manually turning it. (without having to disassemble the drive)


Also, shameless plug. You can get a CF card replacement for the T3100 (made by me) if you really don't want to disassemble the machine.
https://conventionalmemories.com/wiki_cm/3inONEder_for_Toshiba_portables

 

[granzeier]

As for the other suggestions such as idediag, does it even work on non-IDE drives? I'm almost certain that the T3100 has a pre-IDE (MFM?) drive. Apologies if that makes me sound stupid cause I don't know the most about old storage formats.

Apologies from me, I thought that the 3100 used IDE drives. In this case, you can forget about IDEDIAG, no it won't work on non-IDE drives. Sorry.

 

[xjr358]

Also sometimes these drives don't even spin up, or don't get up to speed and speed down. That can usually be fixed by dropping a bit of bearing oil into the spindle motor and manually turning it. (without having to disassemble the drive)

Can you please elaborate this method? Not sure how to turn the spindle manually

 

[Conventional Memories]

Can you please elaborate this method? Not sure how to turn the spindle manually

Of course.

The spindle is on the bottom of the drive.

Use needle pliers to turn the inner nut clockwise (be careful not to damage the grounding tab). Don't have to turn it a lot, just break it free and then power it up.


Best to keep it running for a while to make sure it doesn't get stuck again.
You might have to do this a couple of times if it keeps getting stuck in that spot when it has spun down.

Also a drop of bearing oil in there helps a lot.

I think I have 8 or 9 of these drives working now, but one of them still has trouble spinning up after having been off for a while. So this might not always fix it.

I also have a couple of these JVC drives that have issues with the read head movement getting stuck, but I haven't found a way to fix that yet. This seems to be an issue with most of the 20MB drives.

 

[xjr358]

Got it! Thank you for the detailed explanation! Appreciate. I'll let you know how it turns out.

 

[xjr358]

I replaced all leaked capacitors (don't have an SMD capacitor in stock, temp solution) and oiled the spinner bearing, moved it clockwise but no luck - the drive doesn't want to start the spindle. Maybe some electronic issue. I bought another T1200, will see if that drive will be OK. Anyway - thank you very much and have a great week ahead!

 

[Conventional Memories]

I replaced all leaked capacitors (don't have an SMD capacitor in stock, temp solution) and oiled the spinner bearing, moved it clockwise but no luck - the drive doesn't want to start the spindle. Maybe some electronic issue. I bought another T1200, will see if that drive will be OK. Anyway - thank you very much and have a great week ahead!

that's strange first time i've seen one that doesn't spin up at all. have you tried "starting" it when the T1200 tries accessing it, when the drive indicator led is lid.
maybe a stupid questing but have you checked that the HDD switch on the PSU is on?
If the drive access led comes on and the HDD doesn't spin up after manually starting it I would maybe make sure the power to the drive is good, It has it's own voltage rail and mosfet on the PSU.


Edit: I think I remember having an issue like this in a T1200 where the mosfet was blown because of the high current the disk was drawing to start up with it stuck in place. so that might be the issue here.

 

[xjr358]

Yes, I tried to start it manually, and no reaction - looks like it even not trying to start at all. HDD LED only blinks shortly. The reading head is trying to move but the spindle is silent. I double-checked all voltages from the PSU, and at the HDD controller board everything is fine - there are 12 and 5v. Next week I should get another T1200 - I will try to compare drives and change them between the laptops.

Edit:
Maybe there's a power issue, look how screen invertor 5V is connected - via PJ19 (math coprocessor jumper)
Edit2:
Unrelated. I checked all voltages directly at HDD pcb, there's s two 5V and one 12V - all of them is ok. So will wait another T1200 to proceed.
HDD Pinout:

GridLite HD info (JVC)

 

[Conventional Memories]

Yes, I tried to start it manually, and no reaction - looks like it even not trying to start at all. HDD LED only blinks shortly. The reading head is trying to move but the spindle is silent. I double-checked all voltages from the PSU, and at the HDD controller board everything is fine - there are 12 and 5v. Next week I should get another T1200 - I will try to compare drives and change them between the laptops.

strange, I haven't seen that yet. please post back when you have tested with another drive

Edit:
Maybe there's a power issue, look how screen invertor 5V is connected - via PJ19 (math coprocessor jumper)

That's totally normal, the T1200 wasn't originally developed for a backlight, it only was available without one at the start, and then later they offered the backlight as an option or upgrade. at the end the non-backlight model was discontinued.
The PJ19 is not for the coprocessor, PJ20 is for that. the PJ19 is for an ICE (in circuit emulator) during development and is not used during normal use, that's why they used that to connect the backlight inverter, so that they didn't have to change anything about the motherboard to support the backlight.

 

[xjr358]

Oh, that was the way how they did it at the factory back then! Thanks for the explanation! I thought it was some kind of jump wire that was made during the life of the machine. ICE oooooh that's what it stands for in cirtuit emulator! Anyway, that's not the power issue - I checked all 3 voltages at Hdd's pcb (5, 5, 12v) - will wait for another T1200, I'll update this thread.
BTW
I measured that slot for the hard ram battery and know what? Looks like 14500 battery will fit there perfectly. Will order some batteries and BMS and will report how it works.

 

[3lectr1c]

This JVC drive has additional leaky caps on the back of the PCB unfortunately. The drive is constructed as such that the entire drive mechanism comes out with the PCB, and you cannot separate the PCB from the mech without taking the drive mech apart. Because of this, these drives are basically toast once the caps on the back do their damage.
See this video at 46:30 in:

This may just be the 20MB ones, I don’t know if the 10MB drives have the same caps on the back.

 

[Conventional Memories]

This JVC drive has additional leaky caps on the back of the PCB unfortunately.

Very interesting I haven't disassembled any of these to that point.
I wonder if this is the same on both the 10mb and 20mb models

My success rate for recapping (outside) and unsticking/lubricating spindle are:
10MB: 9/10 working reliably, last one doesn't spin up all the time, still has a sticky spindle.
20MB: 1/5 working reliably. Others spin up but have issues with head movement.

These issues are different from that video.

I guess I'll have to try opening one of the 20MB drives up and take a closer look.

 

[3lectr1c]

The 20MB ones are confirmed to have them, would be more interesting to see if the 10MB ones do. But if you don't have one dead enough that you'd risk opening it, I get it.
For what it's worth though, drives this old don't really get damaged at all from being exposed to outside air, as long as it's clean air.

 

[Conventional Memories]

The 20MB ones are confirmed to have them, would be more interesting to see if the 10MB ones do. But if you don't have one dead enough that you'd risk opening it, I get it.
For what it's worth though, drives this old don't really get damaged at all from being exposed to outside air, as long as it's clean air.

Don't rally have one to to open now. but I will try and see if I can revive one of those 20MB drives now that I know there is more on the inside.
I've done many repairs on 20 and 40MB Connor IDE drives so I have some experience with what not to do. like, don't blow them out with compressed air the residue it leaves is a pain on the platters.