Replacement TI-59 power module help

[Dave Wave]

So the internal power module on my TI-59 went bad. 

It converts the 3.6 DC from the battery to both 10 AND 16 volts DC (two outputs).

Rather than try to trouble shoot the tiny parts on the module, is there some sort of modern DC-DC coverter that will fit into my very limited space (1x1X.25 Inches)?

Any other ideas appreciated.

Thanks,

-Dave

[Psi]

It's hard to spec something without knowing what current you need.

If you want a IC to build your own modern converter this one looks quite good http://cds.linear.com/docs/Datasheet/19441f.pdf


If you want to buy a finished DC/DC unit i doubt there's one that can take 3.6 and output 10 AND 16
but you might be able to do it with two small modules.

This http://search.digikey.com/nz/en/products/NTA0315MC/811-1663-5-ND/1927224 plus a 10v liner regulator would get you 10v and 15v from 3.3.
But it would only be at 33mA total and if your 3.6 is from battery then you might end up driving it out of spec. Its only intended for 2.97 to 3.6V input and your batteries might be 1.5x3 when fully charged.

You might have to use a 2.4V~4V in  --> 3.3V out module and then feed that into something like the above module.

[Dave Wave]

Lets add to the fun:

Current will be around 100 mv

Space is only 1x .5 x .25 inches.

However, I did some poking around on the old board and found what might be the culprit.

The blue cylinder w/ the black band at the top...has a 600 or 800 stamped on it, I think it is a diode, and it shows no resistance either side. just like a piece of wire. Could it be shorted? Is it even a diode?

Thanks!

-Dave

[Psi]

Definitely a diode of some sort. Ya really have to unsolder it to test it though. Unless you're getting open circuit in both directions.

[TerminalJack505]

Someone has posted a PDF of the TI-59 service manual here:

http://bulk.rskey.org/BULK/CALCDOCS/TI/ti59-service-manual.pdf

I gave it a quick glance but it doesn't look like they have a schematic for the power supply module.

Google didn't turn up anything for the supply module part number either.  You might be able to glean some useful information from the service manual anyway.

The part in your picture might be a 1N6000 or a 1N6001.  Those are a 10V zener diode and 11V zener diode, respectively.

If you look at the rest of the diode it will most likely have the rest of the part number on it.

[david77]

I'd also check the tantalums on that board.
It shouldn't be too hard to get that DC/DC converter back to work.

The TI59 is a lovely calculator and easily worth the effort. I've just replaced the battery pack in mine, should be good for the next 33 years  .

[T4P]

I remember seeing many schottky's that look like that and apparently it should be a zener .

[Dave Wave]

Someone has posted a PDF of the TI-59 service manual here:


The part in your picture might be a 1N6000 or a 1N6001.  Those are a 10V zener diode and 11V zener diode, respectively.

If you look at the rest of the diode it will most likely have the rest of the part number on it.

I did some more poking around.

The "diode" is marked 600 on one side and 111 on the other. Possible part #?

The "diode" reads 2 ohms exactly both ways (still in the PCB though).

I checked both tatalums. One gave .8 ohms, but the brown one shows open (nothing) both ways. It is marked +33 then under that +16.

So where to start? Swap out the tatalum and if that does not work try and figure out the diode?

THanks for all the help,


-Dave

[TerminalJack505]

Hmm...   I can't figure out what that part is based on those markings.  There's a picture of the PCB on Wiki and that particular part is reddish-orange.  That's the typical color of diodes so that's probably what it is.  Of course, there's all kinds of different diodes so that doesn't say much.

You might look around the main board and see if you can find the same part.  If you do find another one like it then you can use the service manual to figure out what it is since it has a parts list for the main board.

You might also try swapping out the parts that you can identify and see if that brings it back to life.

As a last resort you can make your own power supply module.  The service manual gives the requirements for the module.  The service manual states that V_dd is (nominally) -10V and V_gg is (nominally) -15.8V.  It's a little iffy on the required current so you might want to just beef it up a little.  200 mA on each rail will probably do.

You can get a couple of inexpensive inverting switching regulators to do the job.  Something like the LM3578, for example.

[david77]

The "diode" is marked 600 on one side and 111 on the other. Possible part #?

Could you have misread the 111? That could well be 1N if the marking's a bit smudged.

[vk6zgo]

Someone has posted a PDF of the TI-59 service manual here:


The part in your picture might be a 1N6000 or a 1N6001.  Those are a 10V zener diode and 11V zener diode, respectively.

If you look at the rest of the diode it will most likely have the rest of the part number on it.

I did some more poking around.

The "diode" is marked 600 on one side and 111 on the other. Possible part #?

The "diode" reads 2 ohms exactly both ways (still in the PCB though).

I checked both tatalums. One gave .8 ohms, but the brown one shows open (nothing) both ways. It is marked +33 then under that +16.

So where to start? Swap out the tatalum and if that does not work try and figure out the diode?

THanks for all the help,


-Dave

If one of your tantalums reads 0.8 Ohms,it is either short circuited,or is connected across something which has gone short circuit.
Pull that tant out & test it out of circuit--if it loses the short,the fault is somewhere else.
The tant with "+33 then under that +16" on it is a 33uF 16V cap.
The + + signs are just to make sure you put the cap in the right way round,they are not meant to be read with the numbers.

[Dave Wave]

If one of your tantalums reads 0.8 Ohms,it is either short circuited,or is connected across something which has gone short circuit.
Pull that tant out & test it out of circuit--if it loses the short,the fault is somewhere else.
The tant with "+33 then under that +16" on it is a 33uF 16V cap.
The + + signs are just to make sure you put the cap in the right way round,they are not meant to be read with the numbers.
[/quote]

Success...sort of.

The Tantalum was shorted...I replaced it...and the power supply seems to work.

One question: This power supply supplies "negative voltage". All ratings are -14v etc..... Positive from the battery is 0 volts.  When I reduce the voltage input, the output voltage goes up. At 4V input, I get 12V output. At 1V input, I get 14V output. This makes no sense to me


I did put everything back in the TI-59 and wonder of wonders it fired up and seemed to work OK....However after about 10mins, the card reader started running on its own (I was not even touching the unit when this started).

I checked both sense switches for the card reader and both seems to check OK.

If I turn the unit on with the CLR button held...I get a 0 with the card reader off....as soon as I let go, the card reader starts up and the display goes blank.

Anybody have and clues to this one or can you point me in the right direction of where I might look?

Thanks for everybodys help TS this issue!!

-Dave

[TerminalJack505]

Happy to hear you're at least partially successful.  So far as your new problem goes, I have to wonder what the power supply module is doing when the card reader starts acting up.  Are the voltages within spec when that happens (as per the service manual)? 

Since there was a short on the power supply module, other components on it may have been damaged.  Is anything on that module getting a little too warm?

I also have to wonder if you are running the calculator from the batteries when this happens or from the adapter?  If you're running from the adapter then are the batteries installed?  Apparently, the batteries have to be installed even when powered from the adapter.

So far as the varying output voltage goes as the input voltage varies: it depends on the specs of the power supply module.  I'd say that so long as the input voltage is within spec AND the output voltage is as well, then there's nothing to be too concerned with.  The service manual mentions the input voltage min and max.  I don't remember what they are of the top of my head.

If you vary the input voltage within the spec and the output is out of spec then that would tell us something.

You might pull the module out and put a couple of dummy loads on each of the rails and let it set and warm up to see what happens to the output voltages and currents.  Use dummy loads that would put each of the rails near their maximum as per the service manual.

Since you have some way of varying the input voltage then you might also see what happens when it is at the minimum and maximum as per the service manual.

I would almost bet that something's still dodgy on the power supply module.

[Dave Wave]

I did check the the voltage levels with the calculator on (while it was malfunctioning), and at least at the test point, they were right on spec.

I put the oscope on the line and there was some noise (but the card drive has a small dc motor in it so I wrote it off to that).

The power supply (unloaded) puts out clean power.

It seems odd that it would run ok for 10 mins, then have issues, then not work again when the unit cooled down again.

What are the odds that one of the transistors could be on its way out?

I guess I could try to hack together another external power supply, hook that up to the unit and see what happens. That would isolate the problem to the calc itself or the PS.


Thanks again for all the help.

-Dave

[TerminalJack505]

Assuming that you can identify them and find replacements, those transistors will be just as easy to replace as the capacitor.  The switching transistor was probably switching straight to ground since that capacitor was shorted-out.

[Dave Wave]

Assuming that you can identify them and find replacements, those transistors will be just as easy to replace as the capacitor.  The switching transistor was probably switching straight to ground since that capacitor was shorted-out.

\

So I pulled the transistors...there are three.

Two were marked QB under that 2104 under that CY. I assume this is a 2104 transistor like a 2sc2104?

The third one is more of an issue. The only clear letters are cs on the top. The numbers are almost totally gone. My best guess is 9013...which does cross to a transistor.

Does this seem to fit with the other two? Would the best bet be to replace the 2104's and install the old "9013".

Any help appreciated.

-Dave

[TerminalJack505]

Assuming that you can identify them and find replacements, those transistors will be just as easy to replace as the capacitor.  The switching transistor was probably switching straight to ground since that capacitor was shorted-out.

\

So I pulled the transistors...there are three.

Two were marked QB under that 2104 under that CY. I assume this is a 2104 transistor like a 2sc2104?

The third one is more of an issue. The only clear letters are cs on the top. The numbers are almost totally gone. My best guess is 9013...which does cross to a transistor.

Does this seem to fit with the other two? Would the best bet be to replace the 2104's and install the old "9013".

Any help appreciated.

-Dave

Making sense of these part numbers is tough.  That's the problem with things manufactured before the internet was invented.

You might use your meter to see if they are, in fact, transistors.  Transistors will just look like two back-to-back diodes so far as when you test them with your meter.

If they are transistors then you might try replacing them with transistors with similar characteristics based on the datasheets and see what happens.

[david77]

I just had a look inside my TI59 to figure out the part numbers.
There's 2x Q8 (QB?) 2104 BY and 1x CS9015 transistors. The diode is a 1N4150.

I think the CS9015 is equivalent to SS9015, PNP universal bjt, replace with BC556 for example.
I have no idea what the parts marked with 2104 might be.
Two diodes are just standard diodes similar to 1N4148.
One diode is marked B2 15, that could actually be a zener?

EDIT: Found this searching for QB 2104, it's a translation from a french forum:

Hello,

This is a transistor in TO92 housing three legs, I deduced that it was a (pnp) and that this was part of a circuit to generate a negative current -10 and -17 V, there are 2 qb2104 transistors in parallel and a cs9105, I replaced it with a 8550 (a qb2104) who walked a bit but after about 30-60min the aircraft became unstable since I have not tried it because I think with 1seul this is to divide the power there are two? looking here and the 8550 is just the amp and is not oscillatory as the 2N3906 that seems present in this type of circuit and it is not the same pattern as there are two 2N3906 crossed but I will liked understand why this transistor is completely untraceable (can be a codification Japanese?) this old circuit was provided and presumably manufactured by Philips in 1980, there is the hand of the circuit number 1013813-1
Thank you, A +

[Chris Kino]

Hi Dave,

I have the schematics, you can PM me if you still want them.

Chris

[Dave Wave]

I just had a look inside my TI59 to figure out the part numbers.
There's 2x Q8 (QB?) 2104 BY and 1x CS9015 transistors. The diode is a 1N4150.

I think the CS9015 is equivalent to SS9015, PNP universal bjt, replace with BC556 for example.
I have no idea what the parts marked with 2104 might be.
Two diodes are just standard diodes similar to 1N4148.
One diode is marked B2 15, that could actually be a zener?

Wow fantastic find! Thanks for the help!

 I'll see what I can do in tracking down these old part numbers.

-Dave

[Dave Wave]

Well nobody seems to know what a qb 2104 is/was. I even had them check some of the old parts catalogs from back in the day and nothing.

Both seem to be PNP transistors. Both seem to check the same on DMM. Is there any way to figure out a modern replacment or am I stuck with throwing these back in?

Second question for the TI-59 fans: The service manual talks about a drop test and the calc going into "pre-load". What is "pre-load".

Thanks,


-Dave

[david77]

Well, as I said I'm pretty sure the CS9015 is equivalent to the SS9015 and you can replace that with a BC556B/C - don't forget to check if the pin out is the same.

Those 2104 devices on the other hand are a bit tricky. Are you sure they are PNP and have you figured out which pin is E, B, C ? Then simply replace them with BC556B/C or with BC327-25/40 for a bit more power.

I think we can safely assume that the transistors in this small DC/DC converter won't be anything special and so any multi purpose PNP transistors will probably do.

[Chris Kino]

I had a quick look on the schematics I sent to Dave Wade yesterday, the transistors are marked QF2014 and QS9015B, the zener is a 15.6V one.

@Dave; I suggest you to contact Joerg Woerner, the www.datamath.org webmaster and send him the link to this thread.
Maybe he can help you out.

[Dave Wave]

Well that was a project...

Here is what I found.

The modern cross to the 9015 transistor is incorrect (x's to a NPN).

1N4148 diode will not work correctly (well the NTE519's I got did not any how)...voltage was about 2volts lower than with the original diodes.

I tried NTE159's to replace all three transistors, but was able to get a bit more voltage out of it by swapping in to original 9015 mystery transistor.

I replaced the zener with a 16volt zener.

The final voltages open circuit were (Vbat 3.97) Vgg 13.64 Vdd 10.2. Both Vgg and Vdd are still a bit low, but the unit runs and selfs tests fine.

Thanks for the help!

-Dave