Toshiba 110cs Rip...

[MAZter]

I am amazed at your persistence, this laptop has no sound, a passive matrix and to top it all off a video card that can't properly scale a resolution any less than 800x600 to full screen, the worst laptop in my collection

 

[Wappynutter]

I know you're right about the 110, that I have left as donor parts. This thread has become the 210cs repair using parts from the dead 110cs really.
Maybe I should have moved the 210 sections and started a new thread but I never did.
The 210 is the bigger brother to the 110cs and has more in common with the 430... dedicated soundcard, better video card etc.. yes still a rubbish screen but it's more about the journey, trial and error, diagnosing and maybe eventually getting it to work again. My slightly obsessive autistic traits won't leave it until it goes bang or catches fire.. but I am real close to just sticking it in a hole somewhere and forgetting I ever owned it. I just feel like I'm real close as I make progress. Today has been a progressive day where I was convinced I'd cracked it... but..
I'll make a post about that soon maybe.

 

[MAZter]

Ok, sorry, in this case I would rename thread.

 

[Wappynutter]

So trace repaired and boom.. no more "ff" code.. Great but I get on boot 3 codes.."00" "01" "00" in that order.
Everytime. So I decided to do as I did on the working 110 and reflowed the crystal..
These pads and solder joints were dull and corroded and happen to sit next to the leaky battery connector and damaged traces.
After a reflow, solder rework I powered up..
I was greeted with codes "00" "80" "81" "80" "00". Great... this is a keyboard error possibly.
Again off and on about 5 times... codes.. "00" "80" "01" "00". Still maybe a keyboard error.
I powered off the board, attached the soundcard, graphics board, external display, powered back up and... "00" "01" "00".
I'm wondering if the extra error codes appeared while the crystal oscillator thing was warm and working but once cooled went back to the usual codes. I have a replacement on the scrap 110 board. I think it's for either the display clock or keyboard clock I'm unsure.
Anyway I've cleaned out some more crusty via's and put the board away for a bit while I plan my next move.
Any thoughts? Thanks again.

 

[sergioag]

I have a Toshiba Libretto 50ct. Yes, it had a small Varta battery inside. Thankfully, I removed it before it caused some disaster. No corrosion at all.
From time to time I'm tempted to put a fresh battery, but so far I've resisted the temptation and I don't want to make any modifications to it. Unfortunately the RTC doesn't work properly without this battery.

 

[Wappynutter]

I loved the Libretto. It was my main win 95 system for about a year in 1997/98 for work. It had a docking station and a modem and little carrycase etc.. I would type up emails and word documents etc.. using office 97 then plug into the phone line using the fax modem to send. Always wanted to get another until I got put off by all the negative reviews over how brittle and easily broken all these years later they had become.
It was a real conversation piece back in the day when out in the field working. How small and compact and well designed it was for the time.
Now days they hold a good price for a clean well maintained example.
Think mine was the 70ct, favorite screensaver was the haunted house. Great memories.

 

[Wappynutter]

Still messing with the 210Cs when I have an hour or so... done lots, tried more but still the same damn codes.. 00, 01, 00.. brought a new lpt port reader incase mine is busted and nope... think the new one is busted as it just stays on one code 00. Even on other laptops...
So, read somewhere I should refresh the bios incase it's messed up.

Chip removed..

Pads cleaned..

Socketed..

Refreshed and installed, checked for activity on the pins.. all working...
No change.. Next option.. large chip reflow..
Mr Thermalwrong advised on a Toshiba post I read that it's risky due to shorts etc.. but I managed to find a loose pin and the solder on all the pins looked dull and soft so......

All the big chips, the Peggy chip and another controller ic..
Cleaned, checked for shorts, powered on and...
00, 01,00... damn...
So checked all the crystals etc.. these two..

This one..

Even this one...

All have correct activity on the oscilloscope, what it says on the chip is what I'm reading activity wise. 4mhz, 2mhz, 32.7mhz, 14.3mhz..
So the only negative it seems is some of the big chips still sit cool including the cpu.
Audio is quiet but I get a pop on power on and pop on power off through the headphone jack. No display either internal or external, no keyboard either internal or external but the external keyboard led's flash when powered up.
So close yet so far... and all just for the practice and experience... and the fact I do like the old grey Satellite's from the mid to late 90's.
Maybe I'll reflow the ram next or recheck the mosfets and power ic's.
It's gonna be something stupid like a resistor or damaged trace/via that's near impossible to find I bet..
Any other ideas or suggestions welcomed.

 

[GiGaBiTe]

Maybe go back and start inspecting the passives to see if there any other bad resistors/cpacitors/transistors/etc.

 

[Thermalwrong]

Hmm, I think I can possibly see two solder bridges on the QFPs?
162/163 of TC160G54JF
238/239 of TC163GB7AF

I'm currently troubleshooting a 400CDT board with power issues and found that after reflowing the TC183G75AF, the laptop went from being able to turn on (and immediately turn off with power fault) to not being able to even turn on when the button was pressed. Took me a while to find two or three solder bridges on that QFP but once cleared the board was back to its original state

 

[Wappynutter]

Thanks for taking a look Thermalwrong, the second I get the iron out I'll check over the big chips again. I'll definitely check those pins you've mentioned.
Good luck with the 400CDT.

 

[Wappynutter]

Just a quick update as I jump between my collection of laptops, base units, consoles and peripherals.
Had a handful of console and controller wins of late so I headed back to the Toshiba pile. The working...ish 110 is no longer starting giving similar codes and behaviour to the 210Cs.
I have been tracing via's for a fellow vcf member and been identifying a failed resistor pack on the 210 in the process..
If anyone can find a start and end point for this resistor in the pack marked in white I'd be very grateful.

I swapped it out for a working pack but still never traced the white marked resistor.

Unfortunately it didn't cure my failed boot up on the 210 so I kept digging again.
I checked the reset pin on the cpu incase it is stuck and that why it sits cold and halted in the boot up, it has a testpad so I probed it.

The 2 bent cpu pins are still attached just dented somehow.


Unsure if to test ac or dc but did both. The ac which is what I decided was the correct one did the expected high then drop low almost instantly. In dc mode it sits high for a good 15 seconds before instant low. Like it was waiting for a start signal that never came so shut off.
I continued to check signals and clock signals from the oscillator/crystals.

this one seemed fine.
Tested these next,
.....and found a problem..
Top of the crystal read the right frequency..

The bottom gives a weird really high signal..

Checked on the ic that recieves or supplies these signals..



The data sheet.

Reference crystal input and feedback.. confusing... anybody got any ideas?
I'm guessing either the crystal is bad or the ic is damaged.
It is after all right next to the damaged battery connector point and corrosion area.
All input and suggestions welcomed..
What's your money on? Crystal or chip?
Or none of the above and this is a correct reading. Anybody know where the pads on the crystal go to?.
Thanks for looking, much appreciated.

 

[Wappynutter]

So maybe progress... probably not...or if it is it will be short lived I'm sure.
So I removed both the crystal and ic that distributes the signals and power.


Cleaned up, cleared the corrosion on the pads and replaced the solder then resoldered the donor components on the board.

After clean up and a cool down I powered the board up.. I expected no real change but...
So I measured the cpu pin on the ic.

Don't know if it's right but I have a good reading at around 5v p2p.
Bottom pins that read crazy before on the crystal measures...

14.2mhz at around 5v p2p and looks consistent.
The upper pins on the crystal measures..

14.4mhz at around 3v p2p.
So I'm guessing these supply my 3v and 5v areas. This is an improvement on my last measuring attempt and while I'm measuring my hand is resting on my cpu heat sink and it's getting hot to touch..
Would ya look at that...

Heat all over the board.. areas have come alive.. have I fixed it?
....of course not..
Still no display or sound or any change.
Got codes... 4 of them..




That is all..
Code 1 flashed and I missed the 01 lit up.
So progress, but not enough to kick-start a display and bios post.
I'll keep going, an hour here... an hour there.. it's all experience and practice.
Any advice or input welcomed. Thanks again, appreciated.

 

[3lectr1c]

Nothing to add in, other than that I have been loving watching your progress as you try to figure this problem out! I hope to see you getting it running soon

 

[Thermalwrong]

That's some serious progress, congratulations You removed the 14.318mhz crystal and clock generator and resoldered them and that seems to have helped? Interesting because I've seen more than one of these boards have corrosion specifically on the clock generator and that would take everything down.

In regards to your new codes, from my experience with the later MMX platform ones, getting stuck on code "04" or going to a random / meaningless code after "04" means that there's a fault with the memory data lines. If there's an addressing or RAS / CAS fault then it'd normally get stuck on code "05" instead.
I fixed multiple bad memory data lines on my own 'board-from-hell' (which took ~8 months to fix), by mapping out all 64 data lines then verifying them. Now I've done that though, the better way is to look around the board for dodgy looking VIAs. While it's off probe the next nearest VIA that looks like its part of the same group so there's a known good signal wire, then brush over all the memory chip data pins to see if you can find which memory chip and pin it connects to, to identify the data lines. Or trace lines off from the known pinout of the memory chips.

The memory data signals on the Satellite 460CDT (MMX platform) are all routed in little groups of 4 or 8 and don't use termination resistors, going straight into the northbridge. Here's what I mean in regards to how the memory data traces are laid out, note this is the later Satellite 460 series which uses a BGA northbridge:


With the 210CS it should be easier to trace out the memory lines because it's got legs instead of BGA pads. The one in the picture above was a very dead example where the corrosion got into the CPU package. Also since there is the maintenance manual for the 420 series (420MM.PDF) that includes the pinout for the memory expansion, you can determine data line numbers & ordering that way.

 

[Wappynutter]

Thanks for dropping by Thermalwrong, greatly appreciated. Your knowledge in all things old and Toshiba is top level when it comes to repairing and diagnosing board level damage. I feel I'm am so close now to having a working 210Cs again.
I have begun tracing these memory vias and traces from the TC203G14JF ic. This is my Northbridge is that correct?
You can follow most traces as like you mentioned they come off of legs and go to memory chip legs or via's.

All down that lefthand side near the memory. Now I have had a couple that go to via's, appear in one place then appear at the topside and into the cpu. Some go to resistor packs and into the memory ic's usually the same leg on multiple ic's, on the front two memory ic's near the bios chip too.
I am struggling to find a start and end point for this resistor marked in white.

The red, yellow and blue marked resistors go to the same pin on multiple memory ic's as mentioned.

All originate off the TC203G14JF chip but no connection found for the white.
Maybe this is my missing trace? I have no working board to confirm.
I shall keep probing and clearing via's if any look crusty and check for continuity.
Thanks again Thermalwrong for taking a look and offering positive info. More soon.

 

[Wappynutter]

Nothing to add in, other than that I have been loving watching your progress as you try to figure this problem out! I hope to see you getting it running soon

Thanks for following along. Hopefully one day, just like my recent IBM repair, it'll all come together and I can play a little lucas arts adventure or two.
More soon.

 

[Wappynutter]

So, like Thermalwrong, I have been busy tracing all the memory via's and traces from the TC203G14JF chip.
Slowly going from hard copies of the connections I have drawn to digital colour marked photos for sharing here.
I have traced and identified start and end points for all the left hand side and some of the lower ones.
I have 33 traces running to the memory expansion slot and cpu and 24 traces running through resistors to the ram ic's.
Well 23.. I still have that 1 trace off that resistor that starts and ends nowhere.
I haven't finished colour marking the cpu and memory expansion slot via's yet but this is the traces finished so far.


So I'm hoping someone can finish this and identify where the orange marked resistor connects.
I'm guessing I have a choice of pin 2 bottom left or pin 4 bottom right of the ram ic's possibly, or maybe not? My checks give me no connections.

This 330 orange marked resistor here(top one in the 4 resisitor pack).
I'll share the rest of the colour marked images detailing the trace start and end points when complete. I'm sure 430 and 210 owners will appreciate the trace knowledge. I did find another bridged resistor so I replaced it but still no difference to start up codes.
More soon, thanks again, appreciate any help from fellow 430/210 owners who can find my missing resistor connections.

 

[markjwyatt]

batteries- you can't live with them, ... you can't live with them.

 

[Wappynutter]

batteries- you can't live with them, ... you can't live with them.

This is true. After 30 years they become parcels of death to all electricals.

So anyway, narrowing my ram chip pins down further I forgot on page 3 of my notes I did this...

So yes, ground connects to pin 4 and bottom pin on the lower right of each ic.
Voltage in connects to the last pin lower left of each ic through a capacitor.
Update the colour marked image.

This leaves just the first 2 pins lower left of each ic. They also don't connect to each other so maybe they are no connects.
Maybe this make the resistor I'm trying to trace a no connect too?
The board was so clean and almost corrosion free, or so I thought.
I'll carry on. Thanks again for all input.

 

[Thermalwrong]

That's the Output Enable line, or OE. On EDO memory that's supposed to be grounded