PCB of garage door opener failed: Which PIC is this?

[Joachim Wunder]

Exactly what markings are on the ?TO220-ish device below the small PCB? If that
is a 3-terminal reg, I'd be checking it out as well.

This is a L7805C-V voltage regulator. Any means to desolder it and
test it offline?

 

[Joachim Wunder]

Hi!


Hmmm...checking the Intel datasheet for the 803x/805x family shows that some
of them (8051/8052/8751) do have onboard ROM program memory. At least one
UV-eraseable version also exists.

Any packaging could be possible, given how many different companies have
produced these parts over the years. Intel mentions a "CERDIP" package in
their datasheet, which I would take to mean "ceramic DIP".

Another thought just occurred to me...and maybe it occurred to you as well.
Did you pull the chip and check the bottom of it to see if there might be
anything printed there?

William

Hi William,

yes, yesterday I once pulled the chip (unfortunately, I already pushed
it in again). It sayed something like a <number>AA, MALAY 963. I did a
reverse lookup on Google, just to find out nothing about it. I think
the <number> simply is kind of a serial number or so. But is there any
site where I could do a reverse lookup of such numbers just to find
out the exact type of the microcontroller? If so, I would dare to pull
the chip again.

 

[Joachim Wunder]

The IC will have custom firmware on it so you won't be able to replace
it. Chances are unless the regulator has failed in a short circuit and
caused the voltage to rise, the chips are probably fine. Check to see if
the filter capacitor is shorted, I recently repaired an old Volkswagen
ECU that had a shorted electrolytic capacitor filtering the 5V output,
resulting in a dead car.

No, the filter capacitor is not shorted, I just measured it. Hmmm.

 

[Joachim Wunder]

It seems to me that this is all getting a bit over-complicated. Why would a
simple device like this even need a micro-controller ? Some years back, it
was fairly common practice to put chips like this onto boards, that actually
did nothing. Literally just there for show to make the item look worth the
money that was being charged for it. Anyway, it seems that there is an
obvious transformer, reccies, and a smoothing cap. Is that where you are
measuring the 12v that you mentioned ? The very hot looking TO220 device
next to the smoothing cap is 99% certain a standard 3 pin monolithic

Yes, it´s a "L7805C-V". Well, I didn´t yet supply any 220V AC voltage
to the board in an offline test yet. I.e. after the garage door opener
failed, I didn´t dare yet to put on 220V~ again. Hmmm, is there any
valid means to test this "L7805C-V" offline with a 12V DC power
source?

regulator IC, and I would guess that it's almost certainly a 5v one, so do

It definitely is a 5v one!

you measure something like 12v, 0v, 5v on its pins ? If you do, do you have
a 'scope ? Can you measure any clock activity at that crystal ? Beyond that,

Unfortunately, I don´t have a scope, only a meter.

without a detailed knowledge of the functionality, and schematics, it's
pretty much impossible to come up with a valid test strategy, that's likely
to result in a repaired board.

Yes, I know. By the way, might he 16-pin IC a I2C maybe? If so, which
ones were commonly used to control an Intel 8748/8749 ?

Arfa

Thanks, Joachim

 

[Meat Plow]

This is a L7805C-V voltage regulator. Any means to desolder it and
test it offline?

Just replace it, they're cheap. Or you could check it in circuit powered
up for 5+ volts between the middle and right pins.

 

[Joachim Wunder]

Hi there,

I have a garage door opener named "Liftboy SMD 3200" which was
installed into my garage back in 1990 (in Germany). I couldn´t get any
information about this garage door opener on the net. The motor
control PCB of it just failed this week. I suspect either any capacity
or the voltage regulator to be the culprit. Does anyone know the
manufacturer of the PCB shown under
http://img255.imageshack.us/img255/9023/cimg0094aq6.jpg ?

Unfortunately, the 40-pin PIC´s type isn´t readable under the sticker
at all ´cause its surface seems to have been scratched off under the
sticker by the manufacturer of the PCB. Yagh. Which PIC is it supposed
to be? There is a Quartz which reads "4.000 SUNNY 7624".

Moreover, the same applies to the 16-pin IC: It´s surface seems to
have been treated by the manufacturer of the PCB, i.e. the exact type
has been completely scratched off on purpose as well. Is there any
way to find out which IC it might be?

I just measured the pins of the 16-pin IC with a logic tester: Pin 14
always stays HIGH at +5V with >2 MHz. Does that point me to a Philips
I2C with a SCL line at Pin 14? If so, which I2C might that be?

Please advise.

TIA,
Joachim

Thanks,
Joachim

 

[Arfa Daily]

No, the filter capacitor is not shorted, I just measured it. Hmmm.

As long ago as this was made, I2C was not a common bus scheme ( if it even
existed at all ) and in any case, what would want to be communicating over
such a bus in a simple door controller ?? The regulator can be checked out
of the board by applying any voltage over about 9v between the left pin (
+9 ) and the centre pin ( 0v ), and measuring the output for +5v between the
right pin and the centre pin. This is with the writing towards you or metal
tab away from you. However, you are never going to get to the bottom of this
board's problems by being afraid of it. A bulb blowing across the primary of
the Tx, is highly unlikely to have done any damage to it, and even if it
had, what is going to be the problem ? An o/c primary winding ? No problem,
the board just won't do anything. A s/c or shorted turns primary ? No
problem - it'll just blow the mains fuse, or smoke. No matter what, it
doesn't work now anyway, and you have no clue why unless you put some power
on, and take some readings. It could be something as simple as a bad power
supply circuit - one of the reccies, or the 5v regulator, or yes, the tranny
even, but unless you build up the courage to put some power on and see just
what's going on, you're never going to know. Anything we suggest at this
point, is really just useless conjecture, yes ??

Arfa

 

[Franc Zabkar]

Yes, I am aware of the fact that I would need a PIC programmer and a
new 8748 or 8749 PIC (I´m still not sure which one it really is).
Pretty good question where to still get such a component from, I
know,... is there any equivalent PIC which offers the same pin
assignment and code compatibility of a 8748/8749? I know, even I doubt
that, but I better ask in this group before giving up.

It could also be an 8751, although the location of the crystal would
suggest a 48 or 49. You should be able to narrow down the likely
contenders for the uC by determining which pins are Vcc (40?), Vdd
(26?), Ground (20?), crystal (2 & 3?), reset (is there a resistor and
capacitor at pin 4?).

FWIW, I have some 8749 chips in my junk box. I also have a device
programmer that can handle them. If you are lucky, and if the uC has
dead ports but accessible EPROM, and if the security bit (?) has not
been set, then I may be able to copy it for you. In the past I've been
able to recover code from bad ROM versions of Intel uCs and replace
them with the EPROM versions.

BTW, I'm in Australia.

- Franc Zabkar

 

[Franc Zabkar]

might the 16-pin IC a I2C maybe? If so, which
ones were commonly used to control an Intel 8748/8749 ?

There appear to be 8 or so interconnect pins at the RHS of the 16-pin
IC. Can you trace them to the uC's pins? Do they go to the data bus,
ie pins 12 through 19? Which pins are power and ground? What is the IC
to the right of the relays? Which chip/component drives the relay
coils?

- Franc Zabkar

 

[Homer J Simpson]

Tried these guys?

http://www.aaaremotes.com/gegadoopcibo.html

 

[me]

As long ago as this was made, I2C was not a common bus scheme ( if it
even existed at all ) and in any case, what would want to be
communicating over such a bus in a simple door controller ?? The
regulator can be checked out of the board by applying any voltage over
about 9v between the left pin ( +9 ) and the centre pin ( 0v ), and
measuring the output for +5v between the right pin and the centre pin.
This is with the writing towards you or metal tab away from you.
However, you are never going to get to the bottom of this board's
problems by being afraid of it. A bulb blowing across the primary of
the Tx, is highly unlikely to have done any damage to it, and even if
it had, what is going to be the problem ? An o/c primary winding ? No
problem, the board just won't do anything. A s/c or shorted turns
primary ? No problem - it'll just blow the mains fuse, or smoke. No
matter what, it doesn't work now anyway, and you have no clue why
unless you put some power on, and take some readings. It could be
something as simple as a bad power supply circuit - one of the reccies,
or the 5v regulator, or yes, the tranny even, but unless you build up
the courage to put some power on and see just what's going on, you're
never going to know. Anything we suggest at this point, is really just
useless conjecture, yes ??

Arfa

Yes. Does the motor work? Do the relays switch? Are all the Diodes
functional? What causes you to suspect the IC's? Are you just whistling
in the dark because the voltage regulator that normally gets hot has
discolored the circuit board it sits under? Without actual
troubleshooting you are not going to get anywhere...

 

[Joachim Wunder]

Tried these guys?

http://www.aaaremotes.com/gegadoopcibo.html

No, I didn´t tried them yet. But I have done some good progress today:
After a whole bunch of further phone calls I was now finally able to
find out the manufacturer of the PCB. It´s a very small company in
Saar/Northern Germany. They are not on the Web, so you simply can´t
find them there, unfortunately. But the company who installed the
garage door opener back in 1990 had a deeper look into their archive
to find out this small company for me, again. Thanks God. Well, even
this company is located in Northern Germany but not presented on the
Web at all. So, life sometimes really sucks, I know.

Well, unfortunately, the techie himself who designed the PCB wasn´t in
today, but another tech person of the same company told me that they
MIGHT have some spare PCB still in stock. Thanks God. But this is only
an assumption and cannot be clarified before the upcoming Monday. So,
I will keep you all posted here. And, moreover, the techie whom I
had on the phone today even told me that all employees are definitely
not authorized to hand out any circuitry plans or specs/types of the
microcontroller and the accompanying IC. Holy cow.... anyway...
there´s some hope the next Monday. I´ll report back and keep you all
posted with an update then.

In the meantime.. thank you ALL so much for your tremendous support!

For the rest of the reponses from all of you, I will answer one after
the other post. So, please be patient with me. I´m just sitting here
with a logical test device to find out any more information.
Joachim

 

[Joachim Wunder]

I just uploaded seperate larger pics of the front and back side of the
PCB:

front side:
http://img184.imageshack.us/img184/3231/cimg0083kj3.jpg

back side:
http://img187.imageshack.us/img187/1555/cimg0084qg6.jpg

I am sorry to not have done that any earlier. On weekends I now have a
little more time to take better pics and start any systematical
diagnostic tests.

There appear to be 8 or so interconnect pins at the RHS of the 16-pin
IC. Can you trace them to the uC's pins? Do they go to the data bus,
ie pins 12 through 19?

As far as I can see Pins 12, 13 and 14 of the microcontroller are
connected to the interconnect pins at the RHS of the 16-pin IC.

Which pins are power and ground?

Pin 8 is V_SS (GND) and Pin 16 is V_DD (+5V) as far as I measured on
the 16-pin IC.

What is the IC
to the right of the relays?

That´s a PHILIPS CNY17-4 Optocoupler.

Which chip/component drives the relay
coils?

That´s actually a BC237B transistor each. They are hidden below the
piggyback PCB, unfortunately.

- Franc Zabkar

Thanks
Joachim

 

[Joachim Wunder]

It could also be an 8751, although the location of the crystal would
suggest a 48 or 49. You should be able to narrow down the likely
contenders for the uC by determining which pins are Vcc (40?), Vdd
(26?), Ground (20?), crystal (2 & 3?), reset (is there a resistor and
capacitor at pin 4?).

I just measured: Pin 40 is Vcc, Pin 26 is Vdd, Pin 20 is Ground, Pins
2 & 3 are crystal. At Pin 3 there is a 22 picofarad capacity as well.
Between Pins 4 & 5 is a resistor with 10kOhms.

FWIW, I have some 8749 chips in my junk box. I also have a device
programmer that can handle them. If you are lucky, and if the uC has
dead ports but accessible EPROM, and if the security bit (?) has not
been set, then I may be able to copy it for you. In the past I've been
able to recover code from bad ROM versions of Intel uCs and replace
them with the EPROM versions.

Thanks for your very friendly offer. I just recovered my old Eprom
Burner from 1992 yesterday night. It´s a German model from www.elv.de
with a separate ISA card, DOS software and an external adapter which
plugs right into the ISA card with a fat ribbon cable. Unfortunately,
right now I only have the 32 pin textool external adapter which is
capable to mainly handly only EPROMs at sizes up to 2 MByte. Anyway, I
just contacted ELV Germany and asked them whether they still have the
external adapter with 40 pin textool socket. This adapter is
explicitely described to be able to handle 8748/49 µCs in the manual
of the burner. And even the old DOS software is able to handle these
8748/49 µCs. Thus, I will see what ELV will respond. I then may get
back to your friendly offer. Thanks a mil.

BTW, I'm in Australia.

Oh, well. I always order most of my satellite dish from Australia.
Thus, distance shouldn´t be an issue for me. *g*

- Franc Zabkar

Kind regards
Joachim

 

[Joachim Wunder]

As long ago as this was made, I2C was not a common bus scheme ( if it even
existed at all ) and in any case, what would want to be communicating over
such a bus in a simple door controller ??

Well, I thought it could be an A/D D/A converter from Philips maybe
(like a PCF8591, for example). Or an I/O expander. But that´s just a
guess, you´re right.

The regulator can be checked out
of the board by applying any voltage over about 9v between the left pin (
+9 ) and the centre pin ( 0v ), and measuring the output for +5v between the
right pin and the centre pin. This is with the writing towards you or metal
tab away from you. However, you are never going to get to the bottom of this
board's problems by being afraid of it.

Well, no reason for me to be afraid anymore: I already applied 230V AC
to the large pins 3 & 4 on the backside of the PCB (counted from left
to right on http://img187.imageshack.us/img187/1555/cimg0084qg6.jpg).
Result: The voltage regulator is definitely not defective at all. I
can read a correct output of +5V between the right pin and the center
pin of it.

A bulb blowing across the primary of
the Tx, is highly unlikely to have done any damage to it, and even if it
had, what is going to be the problem ? An o/c primary winding ? No problem,
the board just won't do anything. A s/c or shorted turns primary ? No
problem - it'll just blow the mains fuse, or smoke. No matter what, it

Well, today I again tracked down in the garage the path where the bulb
was actually connected to: When you look at the backside of the PCB,
PE is connected to the large pins 1 and 2, N(eutral) is connected to
large pins 3 and 8, P(hase) is connected to the large pin 4. The bulb
itself was connected to the large pins 7 (P) and 8 (N), that means
that the current for the bulb is actually switched by the most left
relais when you look at the frontside of the PCB. In consequence, a
shorttime short circuit of the bulb might have sent a current peak via
the most left relais to its +12V control input. The relais itself is
triggered with +12V with this voltage to be switched to the relais via
a BC237B transistor (NPN). As far as I can see from the backside of
the PCB, the transistor itself is controlled by the still unknown
16-pin IC. That´s actually my fear that the transistor and/or the
16-pin IC could have been damaged.

PCB front side:
http://img184.imageshack.us/img184/3231/cimg0083kj3.jpg

PCB back side:
http://img187.imageshack.us/img187/1555/cimg0084qg6.jpg

doesn't work now anyway, and you have no clue why unless you put some power
on, and take some readings. It could be something as simple as a bad power
supply circuit - one of the reccies, or the 5v regulator, or yes, the tranny
even, but unless you build up the courage to put some power on and see just
what's going on, you're never going to know. Anything we suggest at this
point, is really just useless conjecture, yes ??

Right. So, my apologies that I didn´t had enough time before the
weekend to take a deeper look into the door opener in the garage and
onto the PCB again in any more detail. I know I better shouldn´t have
started this thread before having myself more time to do a more
detailed description of the PCB and the way the bulb was connected to
it. Again, my apologies.

Thanks,
Joachim

 

[Franc Zabkar]

I just measured: Pin 40 is Vcc, Pin 26 is Vdd, Pin 20 is Ground, Pins
2 & 3 are crystal. At Pin 3 there is a 22 picofarad capacity as well.
Between Pins 4 & 5 is a resistor with 10kOhms.

I think you're right about the uC being an 8748 or 8749. I'd confirm
that the reset pin (4) is not being held low by a shorted reset cap,
especially if it is a tantalum type. BTW, pin 5 is connected to Vcc,
so the 10K resistor is essentially a pullup.

Thanks for your very friendly offer. I just recovered my old Eprom
Burner from 1992 yesterday night. It´s a German model from www.elv.de
with a separate ISA card, DOS software and an external adapter which
plugs right into the ISA card with a fat ribbon cable. Unfortunately,
right now I only have the 32 pin textool external adapter which is
capable to mainly handly only EPROMs at sizes up to 2 MByte. Anyway, I
just contacted ELV Germany and asked them whether they still have the
external adapter with 40 pin textool socket. This adapter is
explicitely described to be able to handle 8748/49 µCs in the manual
of the burner. And even the old DOS software is able to handle these
8748/49 µCs. Thus, I will see what ELV will respond. I then may get
back to your friendly offer. Thanks a mil.

I have an old (1991?) Sunshine Expro-60 with an ISA controller card in
an old 486 box.

Oh, well. I always order most of my satellite dish from Australia.
Thus, distance shouldn´t be an issue for me. *g*

That's a big surprise. I would have thought these dishes were cheaper
in Europe where satellite TV is much more widespread.

Kind regards
Joachim

- Franc Zabkar

 

[Franc Zabkar]

I just uploaded seperate larger pics of the front and back side of the
PCB:

front side:
http://img184.imageshack.us/img184/3231/cimg0083kj3.jpg

back side:
http://img187.imageshack.us/img187/1555/cimg0084qg6.jpg

I am sorry to not have done that any earlier. On weekends I now have a
little more time to take better pics and start any systematical
diagnostic tests.


As far as I can see Pins 12, 13 and 14 of the microcontroller are
connected to the interconnect pins at the RHS of the 16-pin IC.


Pin 8 is V_SS (GND) and Pin 16 is V_DD (+5V) as far as I measured on
the 16-pin IC.


That´s a PHILIPS CNY17-4 Optocoupler.


That´s actually a BC237B transistor each. They are hidden below the
piggyback PCB, unfortunately.


Thanks
Joachim

AFAICT, the 16-pin IC, if the piggyback PCB faithfully mimics the
pinout, has at least three outputs at pins 1,2,3, each of which drives
the base of a relay driver transistor. Each pin has a pullup resistor
to Vcc, so I suspect that the outputs may be open collector.

Since pins 10,11,12,13 of the IC are connected to the uC's data bus, I
suspect that they may be inputs. If so, then I believe that these
inputs would need to be latched. The fact that there is a "VOH" on the
body of the chip suggests that it might be programmed, in which case
it could be a PAL.

- Franc Zabkar

 

[Arfa Daily]

Well, I thought it could be an A/D D/A converter from Philips maybe
(like a PCF8591, for example). Or an I/O expander. But that´s just a
guess, you´re right.


Well, no reason for me to be afraid anymore: I already applied 230V AC
to the large pins 3 & 4 on the backside of the PCB (counted from left
to right on http://img187.imageshack.us/img187/1555/cimg0084qg6.jpg).
Result: The voltage regulator is definitely not defective at all. I
can read a correct output of +5V between the right pin and the center
pin of it.


Well, today I again tracked down in the garage the path where the bulb
was actually connected to: When you look at the backside of the PCB,
PE is connected to the large pins 1 and 2, N(eutral) is connected to
large pins 3 and 8, P(hase) is connected to the large pin 4. The bulb
itself was connected to the large pins 7 (P) and 8 (N), that means
that the current for the bulb is actually switched by the most left
relais when you look at the frontside of the PCB. In consequence, a
shorttime short circuit of the bulb might have sent a current peak via
the most left relais to its +12V control input. The relais itself is
triggered with +12V with this voltage to be switched to the relais via
a BC237B transistor (NPN). As far as I can see from the backside of
the PCB, the transistor itself is controlled by the still unknown
16-pin IC. That´s actually my fear that the transistor and/or the
16-pin IC could have been damaged.

PCB front side:
http://img184.imageshack.us/img184/3231/cimg0083kj3.jpg

PCB back side:
http://img187.imageshack.us/img187/1555/cimg0084qg6.jpg


Right. So, my apologies that I didn´t had enough time before the
weekend to take a deeper look into the door opener in the garage and
onto the PCB again in any more detail. I know I better shouldn´t have
started this thread before having myself more time to do a more
detailed description of the PCB and the way the bulb was connected to
it. Again, my apologies.

Thanks,
Joachim

No problem, at least we know what's going on now. I still would be reluctant
to believe that the bulb is going to have done anything directly to the
board, evn if it is connected to it by way of being relay switched. The
relay contacts are electrically isolated from the coil by virtue of
substantial mechanical separation. Now that you've got as far as putting
power on the board, you should be able to look for relay drive activity, by
using the test button, or by fooling the board into thinking that it's
installed.

Arfa

 

[Joachim Wunder]

No problem, at least we know what's going on now. I still would be reluctant
to believe that the bulb is going to have done anything directly to the
board, evn if it is connected to it by way of being relay switched. The
relay contacts are electrically isolated from the coil by virtue of
substantial mechanical separation.

That´s what I was convinced too up to the day when the bulb was blown
for the first time. Well, I´ll tell you something, I have one sheet of
paper which came with the PCB, but which only describes part of the
external connectors (i.e. actually only the ones which interface to
the motor on the backside of the PCB). The "funny" thing about is that
this sheet shows a very rough outline of the PCB´s external connectors
and mentions a fuse which is said to be there in the (N)eutral line of
the PCB´s power supply. Ha! I would be glad if there was that fuse
which is rated at 1A F(link). "F(link)" in German means with very
quick response time. And what do I see when I look at the backside of
the PCB? Ha! The manufacturer has simply left out the fuse and has
soldered a piece of wire instead of the fuse holder. Well done.

Now that you've got as far as putting
power on the board, you should be able to look for relay drive activity, by
using the test button, or by fooling the board into thinking that it's
installed.

That´s not easy to do, ´cause I have no documentation of what signals
the RHS large pins on the frontside of the PCB wait for. So, putting
the PCB into thinking that it´s installed is not (yet) possible for
me. I only know that the external remote control receiver is connected
to two of these pins. Moreover, two external limit switches are
connected to the other 4 pins. I first would have to find out more
details measureing the switches out in the garage.

Arfa

Again, thanks for your kind support.
Joachim

 

[Jamie]

That´s what I was convinced too up to the day when the bulb was blown
for the first time. Well, I´ll tell you something, I have one sheet of
paper which came with the PCB, but which only describes part of the
external connectors (i.e. actually only the ones which interface to
the motor on the backside of the PCB). The "funny" thing about is that
this sheet shows a very rough outline of the PCB´s external connectors
and mentions a fuse which is said to be there in the (N)eutral line of
the PCB´s power supply. Ha! I would be glad if there was that fuse
which is rated at 1A F(link). "F(link)" in German means with very
quick response time. And what do I see when I look at the backside of
the PCB? Ha! The manufacturer has simply left out the fuse and has
soldered a piece of wire instead of the fuse holder. Well done.

Fuse links don't need to be actual fuses that you recognize, they can
be special wire used, what looks like a resistor on the board or simply
a small foil path that is 2 half-moons with a small link of foil between
them..