Need some help repairing an 80's laptop PSU board

[FuST]

I'd like to get some more knowledgable help on my current project as I seem to be stuck.

I came across a 1988/1989 Toshiba T1600 on my local ebay equivalent for a pretty good price so I picked it up.
As with most 80's Toshiba equipment I fully expected to find a leaky backup batteries and leaky caps and hoped that would be all but unfortunately this machine has some deeper issues.

The TL;DR is that the RAM backup battery had exploded and corroded most of the I/O shield and part of the PSU board.
Some caps were leaking and I found 2 dead MOSFETs, a dead SMD tantalum and a completely disintegrated SOT32 marked as 'Q29' that has lost it's markings.
I've replaced the caps, MOSFETs and tantalum but somewhat unsurprising the PSU doesn't work.

After spending days tracing out the circuit and have come up with a (somewhat messy) schematic to try and help me figure out what could be wrong it turns out it's basically a bunch of buck/boost converters controlled by 2 Toshiba microcontrollers.

I'd like to know what Q29 could possibly be. I'm thinking it must be a transistor as all other components marked 'Q**' are transistors but  I'm unsure if it's NPN or PNP and what it's orientation/pinout is.
It's included in the schematic but should not be assumed to be correctly oriented.
I would also like to know if I'm guessing correctly that the whole thing (except for the N9V and LCDV circuits) doesn't work unless Q14 is driven.

The UPA1600 IC's (IC5 and IC11) as well as the T4049BF (IC6) tested OK.
I've replaced the ceramic resonator X2 (CSU500p), the LP324's (IC& and IC8), all electrolytics (with low-ESR Panasonics), the tantalum C28, Q4 (2SK814),Q2 (2SK1059) and Q14 (2SJ132)

Right now I'm thinking one or both of the microcontroller might be damaged

Attached are the PNG of the schematic and 2 pictures of the board.

Some links that have been helpful:
US patent application for the PSU logic: https://patentimages.storage.googleapis.com/24/4b/7b/25d6e03be718e9/US5383140.pdf
US patent for the battery charge logic: https://patentimages.storage.googleapis.com/fa/c7/e2/4c02829354891c/US5300874.pdf
US patent for the battery-powered system: https://patentimages.storage.googleapis.com/7d/e9/e5/fd65858366b1ec/US5553294.pdf
Toshiba T1600 Maintenance manual: http://www.minuszerodegrees.net/manuals/Toshiba/Other/Toshiba%20T1600%20-%20Maintenance%20Manual.pdf

Pictures: (Forum won't allow me to upload the files)
Board bottom side
Board top side
Partial schematic

[fzabkar]

Q29 is the switching element in a buck-boost converter. It generates a boosted supply voltage at C6. Q27 shuts down Q29 if the output voltage exceeds the zener voltage of diode 9. Q14 is the on/off switch which enables/disables this boost converter.

That said, the way the circuit is drawn means that Q29 is always on, in which case it would be quickly overwhelmed by the current through a saturated L6. This means there must be an alternative circuit driving the base of Q29 with PWM pulses.

IMHO Q29 is correctly drawn as an NPN transistor (or possibly an n-channel MOSFET). It should be a high speed, low sat type. The source of the base/gate drive is IC13-P34 which is buffered by IC5 and output on pin 18. I would remove Q29 and observe IC5-18 with a scope. You should see a PWM signal. Otherwise, if this pin is stuck high, then this would imply that IC13 is brain dead and would also explain why Q29 failed.


Edit:

You might find this datasheet repository to be useful:

https://mirrorservice.org/sites/www.bitsavers.org/components/toshiba/_dataBook/

[FuST]

That's very helpful, thank you. Have been scratching my head for a good while about this.

Upon (even)further investigation of the PCB there were 2 broken traces, one for the voltage devider/sense network for the batteries (R135) and a pullup for a line on IC13 (R137).
Fixing those got it to at least show some signs of powering up so with that and your explanation it should be easier to further diagnose.

Edit: I just noted there might be an error in your explanation which could have some implications. Diode 9 is a Shottky diode, not a zener.
Would that mean that Q29 would be connected differently? I just drew it in in a way that made sense to me, it could well be that it is in fact very different but I have no way to check that.

[fzabkar]

The only diode which makes sense to me in that position is a zener (or maybe a TVS diode). If it really is a Schottky rectifier, then I have no idea how it works.

I found this tutorial and animation to be very helpful:

Boost Converter Operation:
http://www.learnabout-electronics.org/PSU/psu32.php


Edit: I can't locate IC14 in your photographs, but I suspect that it is a 7404 inverter, or maybe 7414.

Something like this ...

https://www.ti.com/lit/gpn/SN74LVC1G04-EP


Edit #2:

TC7SU04F, TC7SU04FU, Toshiba, high speed C2MOS inverter, 2V - 6V, marking E6, SSOP5:
http://www.mouser.com/ds/2/408/TC7SU04F_datasheet_en_20140512-738695.pdf

[FuST]

That's the part I got stuck on, might well be that some other components in that circuit got damaged and are something else entirely.
However, as far as I can tell, all components marked without prefix and a box on the silkscreen are diodes or dual diodes in various configurations so it would make sense that "7" and "9" are as well.

I'm familiar with how boost converters operate but it is just beyond me how this would make any sense. I'm looking to get a second T1600 so I can cross-check as this particular one seems to be quite damaged.

I'm amazed you found that Toshiba datasheet, that matches IC14 exactly. I've spent hours searching the internet and SMD codebooks and came up empty, even knowing that it should be some kind of inverting component.

After fixing the broken traces in my last post I did fix a couple of weird gremlins:
The right battery's charge indicator isn't always on, makes sense (pun intended) as the sensing voltage devider wasn't connected.
Also the PSU tries to power on briefly but then shuts down in "safety mode" and I need to wait for about 30 seconds before retrying.

Q14's gate is definitely being pulled low, causing it to conduct and powering C7/C6.
VCC (+5v) is supplied briefly to the system, the indicator LEDs flash on briefly, implying the system is trying to boot.
For some reason I can't really explain Q2 blew once and after replacing overheated but I'm not able to reproduce it in any reliable way.
This makes me beleive there is a short to ground somewhere in the 12V circuitry causing the MCU to shut everything down.

Might have damaged some other components when Q2 blew so more desoldering and taking measurements will be needed.
Can't wait to upgrade my ancient analog scope to a storage scope so I can actually see what's going on instead of just a short blip on the scope or Fluke 27

[fzabkar]

This is a good datasheet resource:

http://chip.tomsk.ru/chip/ChipDoc.nsf/vc1?readform&start=1500&view=smd&cat=E&count=500

[FuST]

Thank you for that link.

I got a second T1600 yesterday, It's the newer T1600/40 model with a slightly newer revision PSU board but most components are the same, just laid out a bit differently.
On this board I could just barely read the markings on diode "9" and dsing the datasheet resource you provided I was able to find it: https://datasheetspdf.com/pdf-file/1259546/Toshiba/02CZ18/1.
Indeed, it is a 18V zener diode.

Q29 was good on this board, according to my component tester it is a N-channel MOSFET with a 2.0V threshol voltage and a RDS of 4.8 Ohms. That RDS seems very high but that could just be the Chinesium in the tester that is throwing things off 

Using that information I was able to complete the schematic for the first PSU board, now on the the second board. Hoping that one can be repaired more easaly to cross-check signals. It seems way less damaged by corrosion so I'm hopeful it's still good.

[fzabkar]

Here is another good datasheet resource:

http://www.smdmark.com/en-US/search/code?id=18x