Help, battery drainage and the Zener diode on a HP48G

[Julio]

Hi everyone!! My name is Julio and this is my first time on these
forums, I'm an electronics engineer, and I'm trying to solve the Hp48G
battery drainage problem.

Does anyone have had this issue? My batteries, no matter if new
alkaline or rechargeable Ni-CD, drain in 8-10 days even if the
calculator is off. I get a "low batt" mesage after a few days, I
measure the batteries and always the center battery is empty.

I found some references on the Internet about this problem. I saw some
schematics, and it seems that the problem might be the Zener Diode
Z3, 5.6 volts. It belongs to the power-supply sub-circuit. It is
placed in parallel with the battery. If you place the batteries
reversed, it simply fries.

I managed to open my calc (in a painful procedure), and I tried to see
the diode. I found 4 orange diodes, but I think they are normal diodes
that should be left untouched. But I noticed a small 3-pin black
component, labeled 5A63, placed in the upper left side of the printed-
circuit board (I have pics, if someone wants them, please let me
know).

I have a couple of questions...can some experienced user help me?. My
questions are:

1. How do I know that my diode is fired and needs replacement?
2. Why this diode has 3 pin connections? I found the positive and
negative ones, but the third one puzzles me...
3. It is enough to remove the Zener very carefully? Should my
calculator be protected against voltage peaks if I do so?
4. It should be better if I replace the Zener with a 2-diode
combination? (I mean, normal Zeners and signal diodes bought in a
retail store).

Plan A. Remove the Zener and that's all..
Plan B. Soldering small wires to light-blue coil L1 (+) and black
capacitor 1000uF (-) in order to get connections for my diodes.
Plan C. Placing 2 5.6V (or 5.1V) Zeners in series, but in opposite
directions [in parallel to the battery]
Plan D. Placing 2 diodes in series, a Zener (5.6 or 5.1) in series
with a normal signal diode [but in opposite directions]. The idea is
to use the Zener as usual, for preventing voltage peaks, and the
normal diode for blocking any current when the battery is reversed.

Please, can someone give me some tips? I would like to proceed wisely
and repair my HP48G once and for all...It is some way to contact Mr.
Dave Arnett and ask about this? (Maybe he could be so kind to provide
me some answers), If someone wants pics and diagrams, please let me
know.

Sincerely yours.
Julio.

[Han]

Plan A is sufficient. Most of the later HP48 calculators do not even
have this diode installed.

Han

[Julio]

HEEEEEELP!!!!

I removed the black 5A63 small "Zener" component...and my calc crashed
thereafter:( (I removed with a low.power solder iron and a
desoldering pump, very careful, thinking that everthing would be
fine).

When I turned it on, I saw just single horizontal or vertical
lines....and I got nervous...

1. Did I removed the wrong component?

In some 48GX calcs, I saw the small black component at the left of the
light-blue coil L1. (In my case, this was missing)
In my 48G, the black component was placed at the upper left corner,
just below the EPSON chip. Because it was in parallel with the battery
(and reading 5.6A), I removed it....

Third pin was mysterious...it was soldered but the board track was not
obvious....I removed the component and it seems there was "diodes"
going from this pin to the other 2 pins...

2. Does my calc needs reset? (To use the reset pin, I need to assemble
everything again...)
3. Should I place the component back to its original place?
4. Did I removed the right thing and I should just wait some time?

Please, if you have some expertise, can you help me to calm down and
try to repair my calc from disaster?

Sincerely yours.
Julio.

PD. Should I check also current consumption?
Before this, I measured and I got 8-10 mA for normal usage, and up to
30-35 when doing complicated tasks or operations.
When the calc was off, there was a small microapere current...
But I had that issue, central battery draining, in 8-10 days...

[Han]

Try a reset first. (use the reset pin underneath one of the rubber
feet).

Han

[Julio]

Thank you very much Han for your advice!!! :)
Does anyone in the forums can help me to identify the Zener? (I hope
I have removed the right stuff, and that I had not taken a transistor
away...I have pics, but I do not know how to post them here).
Greetings.
Julio.

[Julio]

Hi everyone!!!
I'm still trying to think what possible went wrong...with this Zener
issue.

I hope I can resurrect my calc...

I have at the moment one major question, which is essential for me.
How can I know if I removed the right component, or if my calc maybe
have never had the Zener diode at all?

I hope someone could give me some hardware tips. Are minor differences
with the diode locaton between early, HP48G, HP4GX, and it may be
missing on late HP48Gs..?

I ask this because some other used reported he removes the componenet
labeled Z3G and everthing went fine..
A. My calc had not this Z3G (near coil L1, light blue sphere)...the
space was empty,
B. I removed instead 5A63, parallel to the battery (some calcs do not
have this 5A63)...

I would try to reset the calc, and if nothing happens, reinstall
5a63..but I think I will wait to get more info...

Greetings.
Julio.

[Han]

http://www.hpcalc.org/hp48/pc/pictures/hp48top.jpg

Here is a schematic for the HP48:

http://www.hpcalc.org/details.php?id=3543

Han

[Julio]

Dear Han.
Thank you very much for your help and pic!!

A. MISSING ZG3 .

My calculator is identical to the one in the picture you posted. If
you notice, on the bottom, there are 3 empty contact points (3 silver
dots) just at the left of coil L1 (light-blue sphere). I think this
was the original location for the Z3G component (this was a simple 3-
pin Zener, one of the pins has no use and it is simply connected to
his neighbor).

B. THE 5A63 COMPONENT.

In my calc, the ZG· component is missing. But if you notice, on the
left side,near the edge of the board, there is a row of small black
components. The upper component of that row is a 3-pin black
component, 5A63 (on the pic, it has a strange golden dot). This is the
one I removed!! (I had not the full information then).

The component can be seen very clearly on this pic (upper left
corner).

http://busylog.files.wordpress.com/2011/04/5.jpg

I think this was a transistor with a different pin configuration (I
got 3 diode measurements among the pins). I think I will solder it
back again (and make some praying!!).


C. THE ZG3.

Here there is some interesting picture about another HP48 calculator
(I think this particular model may be earlier). It has the ZG3
component near the light-blue coil. This component can calso be found
on the the schematics you gave me. But if you notice, these
calculators have the 5A63 missing....

http://www.finetune.jp/~lyuka/interests/calc/hp48g/hp48g-pcb.jpg

I'm thinking now that ZG3 and 5A63 may be totally different
components. I do not know if 5A63

[Julio]

¡I'm thinking now that ZG3 and 5A63 may be totally different
> components. I do not know if 5A63 is some kind of transistor or JFET or diode pack, but it seems that it replaced (and maybe improved) the functions of ZG3. (i think designers just removed ZG3 and found some space on the upper left corner to install 5A63).

5A63 is missing on the schematics you gave me.

Thank you, and if someone has further information, please let me know
(I'm trying now to revive my calc, I hope I could still make it) ;)

Greetings from Guatemala.
Julio.

[Julio]

Hi everyone!!
Good news!!! My calc resurrected...after soldering back the 5A63 (and
making a reset with an alligator clip, making a bridge from the
negative contact to the reset contact, an oversized golden contact on
the upper right, near the edge, I found the right position measuring
carefully with a caliper)...the calc was bring back into life. (but
the whole procedure make me nervous).

Summary:
1. Some calcs have the ZG3 and some others a different 5A63 component.
You just realize until you compare pictures of both models side to
side. ZG3 seems to be not essential to the calc operation, but 5A63
does and it is related to the power circuit.

2. Opening the calc is a very painful procedure.Making repairs is
delicate...if you made mistakes, you can forget about making
calculations with complex numbers and graphics for a while with this
wonderfl equipment...until you get the HP50G.

3. Current consumption,
In my calc, was around 5-8 mA when the calc is ON.
It rises to 30-35 mA when the calc is making complicate operations.
it is around 500 uA (0.5 mA) when the calc was off.

4. The center battery gets drained somehow.... (due to the constant
500 uA?) the best solution that I sugget (if you have no important
things in memory) ist simply to remove it. Battery drainage is an
issue on these calculators.

I hope power issues are better in the HP50G, I think current
consumption should be very low when the calc is off, and there should
be other possibilites, like getting supply from an USB port, or the
possibility to connect a small adapter or power supply.


Closing was not so painful as the opening, but I followed Han's
tutorial...by making the side metal latches 45 degrees with a small
screwdriver, ...straightening the plate, softening the rivets a little
bit...the rivets are a trouble, but I managed to press them very
carefully with some medium-size pliers...

Opening the calc left several scratches on the plastic. I removed all
of them with some file tool....(put some tape film to protect the
plastic in front of the IR ports).

Greetings.
Please, any further technical information, let me know.

[ba...@barend.plus.com]

On Friday, May 11, 2012 3:29:16 PM UTC+1, Julio wrote:

>
> The component can be seen very clearly on this pic (upper left
> corner).
>
> http://busylog.files.wordpress.com/2011/04/5.jpg
>

From the picture the marking is 5A6. This could well be a PZM5.6NB2A, a back-to-back 5V6 zener (datasheet: http://www.nxp.com/documents/data_sheet/PZM_NA_SERIES.pdf)

So, being an electronics engineer you have a multimeter and measured from the device to the battery to ensure it is in parallel to the battery to ensure it is a candidate for removal (it's quite for from the battery side). Also you checked that there were obvious low impedances in the device, and once removed checked the working of the device with the help of resistors and a power supply. So before starting to remove stuff, you looked at what could be happening by connecting the three AAA (and not a power supply) to the opened claculator to see where the drain could be? (Measure where the power goes and where does the voltage significantly drop or perhaps a leaky capacitor is to blame).

It's likely not an easy task, but you have to go back to basics and first identify candidate components and devise ways to test them. You may even find out it's an IC that's impossible to obtain.

Regards.

[Julio]

Dear Barend.
Thank you for the information about the double Zener pack!! I didn't
connected the power supply to the calc, [I found that somewhat
dangerous], so I placed the batteries on their compartment and I
connected everything with alligator clips). That's how I measured the
current, to have an idea what was going on.

I just read that the diode was not any of the 4 orange diodes on the
board, but some small 3-pin surface.mounting component....but I
realized I have done a mistake when another user described it as
having the label "Z3G" and having 2 pins short-circuited. (I was
searching for a 5.6V Zener diode, so when I saw "5A63" (the one in the
pic is 5A6-6) and found their connections and its "diode-like"
behavior, I just removed it (I thought it was an easy "piece of cake"
operation")

... I don''t know why, but my calc simply didn't worked when I removed
this device (and I never found where pin 3 was connected to).

-

1-
|
¬
^
------3
v
-
|
2+


I could not test 5A63 in a protoboard,,the device is tiny and their
connections are too small to be connected to something, but I managed
to measure with the tester. I got a diode behavior when measuring form
pins 1 to 2 (!?)...on the other way around, I found nothing (I'm not
sure if my Zeners are 100% OK), but measurements from 3 to 1 and from
3 to 2 agree with the diagram.

I really would like to know where pin 3 is connected to, and if it is
possible to replace the device with a couple of 5.6V Zeners from the
retail store. But these procedures are really too risky....(board
tracks are fragile when soldering or desoldering), static electricity
is another issue, connecting batteries with alligator clips is
somewhat clumsy, and resurrecting calculators is not a nice
experience)...but I hope that maybe one day, we will have a formal
description about the HP48G power supply circuit and basic
measurements.

Greetings and thank you for the tips.
Julio.

[Julio]

HI!! I found some general information about Zener circuits with double
diodes, used basically as voltage clippers.

http://www.electronics-tutorials.ws/diode/diode_7.html

What should a multimeter measure from pins 1 to 2 (with the diode
function ->|-) I asumme that nothing...

In my component 5A63, I measured a "diode" from 1 to 2 (a 0.6 voltage
drop), but checking the battery + and - signs, I think it will only
conduct current if I place my batteries backwards.
(From 2 to 1, I measured nothing)...



Batt -

1

|
¬
^
|
------- 3
|
v
-
|

2

Batt +

Do you think there is a chance that my component is leaky? Do you
think that theoretically, this 5A63 could be replaced by external 5.6
Zener diodes? (In practice, I do not want to touch any of these
components anymore until I'll be totally sure of what I'm doing...this
component was not present on the schematics that you can get at hpcalc-
org...and as I mentioned before, some calculators do not have it, they
have ZG3 instead...

Where can I find more information about the HP48G power supply
circuit? I think the circuit is somewhat interesting...

Greetings.
Julio.