I've been given a Mercedes-Benz instrument cluster where a flex cable
has come adrift due to mishandling. The PCB is microcontroller based
and interfaces with an odometer via a flex cable. The other gauges
connect via long pins. AFAICS, the flex cable is attached with some
kind of glue - there are no clamps, no screws, no rubber strain
relief. The traces have a very fine pitch, and the cable is almost too
short to work on. The PCB is single sided, so I could probably drill
two holes and add a clamp with a rubber sandwich, but I need to ensure
that the copper traces make reliable contact. Any suggestions?

BTW, the mechanic that gave me this job was doing a freebie for a
friend. AFAICT, he was trying to cover the empty lamp socket of an
annoying glow plug indicator with black tape.


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

"Franc Zabkar" wrote in message
...
| I've been given a Mercedes-Benz instrument cluster where a flex cable
| has come adrift due to mishandling. The PCB is microcontroller based
....
| BTW, the mechanic that gave me this job was doing a freebie for a
| friend. AFAICT, he was trying to cover the empty lamp socket of an
| annoying glow plug indicator with black tape.

What did the indicator light say? "Don't disassemble this"?

N

On Mon, 15 Nov 2004 14:53:39 +1100, Franc Zabkar
wrote:

I've been given a Mercedes-Benz instrument cluster where a flex cable
has come adrift due to mishandling. The PCB is microcontroller based
and interfaces with an odometer via a flex cable. The other gauges
connect via long pins. AFAICS, the flex cable is attached with some
kind of glue - there are no clamps, no screws, no rubber strain
relief. The traces have a very fine pitch, and the cable is almost too
short to work on. The PCB is single sided, so I could probably drill
two holes

Are you very sure ?
'Very fine pitch' and 'single sided PCB' sound like a
contradiction to me. (Or as a euphemism for 'multi layer :-)

--
Kind regards,
Gerard Bok

On Mon, 15 Nov 2004 03:58:46 GMT, "NSM" put finger to
keyboard and composed:


"Franc Zabkar" wrote in message
.. .
| I've been given a Mercedes-Benz instrument cluster where a flex cable
| has come adrift due to mishandling. The PCB is microcontroller based
...
| BTW, the mechanic that gave me this job was doing a freebie for a
| friend. AFAICT, he was trying to cover the empty lamp socket of an
| annoying glow plug indicator with black tape.

What did the indicator light say? "Don't disassemble this"?

What it should have said was "CAUTION: PCB obscures very poor design".

I do quite a few jobs for this mechanic and find that he and his staff
are very competent. Unfortunately I don't have the unit with me,
otherwise I would post a photo so you could see it for yourself. As I
said in my OP, there is no strain relief for the flex cable, and no
connector. The printed traces of the cable and the PCB rely purely on
a tiny, almost imperceptible dab of glue to keep them in contact. I've
never seen anything so abominably bad. Even children's toys are better
built.

BTW, this abortion is manufactured by Magneti Marelli.


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

"Franc Zabkar" wrote in message
...

| What it should have said was "CAUTION: PCB obscures very poor design".
|
| I do quite a few jobs for this mechanic and find that he and his staff
| are very competent. Unfortunately I don't have the unit with me,
| otherwise I would post a photo so you could see it for yourself. As I
| said in my OP, there is no strain relief for the flex cable, and no
| connector. The printed traces of the cable and the PCB rely purely on
| a tiny, almost imperceptible dab of glue to keep them in contact. I've
| never seen anything so abominably bad. Even children's toys are better
| built.
|
| BTW, this abortion is manufactured by Magneti Marelli.

Is that Italian for Lucas Electric?

http://pw1.netcom.com/~krk/lotus/humor/lucasjokes.html

http://members.tripod.com/~tcotrel/lucas.html (think about it)

NM

On Tue, 16 Nov 2004 07:02:18 +1100, Franc Zabkar
wrote:

On Mon, 15 Nov 2004 03:58:46 GMT, "NSM" put finger to
keyboard and composed:


"Franc Zabkar" wrote in message
. ..
| I've been given a Mercedes-Benz instrument cluster where a flex cable
| has come adrift due to mishandling. The PCB is microcontroller based
...
| BTW, the mechanic that gave me this job was doing a freebie for a
| friend. AFAICT, he was trying to cover the empty lamp socket of an
| annoying glow plug indicator with black tape.

What did the indicator light say? "Don't disassemble this"?

What it should have said was "CAUTION: PCB obscures very poor design".

I do quite a few jobs for this mechanic and find that he and his staff
are very competent. Unfortunately I don't have the unit with me,
otherwise I would post a photo so you could see it for yourself. As I
said in my OP, there is no strain relief for the flex cable, and no
connector. The printed traces of the cable and the PCB rely purely on
a tiny, almost imperceptible dab of glue to keep them in contact. I've
never seen anything so abominably bad. Even children's toys are better
built.

BTW, this abortion is manufactured by Magneti Marelli.


- Franc Zabkar

I don't like your chances Franc. The only thing I an think of is
electrically conductive epoxy.

"Ross Herbert" wrote in message
...

| I don't like your chances Franc. The only thing I an think of is
| electrically conductive epoxy.

If it's a zebra connection wouldn't that be worse? You'd short everything to
everything.
He could just try to align it as best he can and hold it there to test. Glue
it if it works.

N

On Tue, 16 Nov 2004 02:03:45 GMT, "NSM" wrote:


"Ross Herbert" wrote in message
.. .

| I don't like your chances Franc. The only thing I an think of is
| electrically conductive epoxy.

If it's a zebra connection wouldn't that be worse? You'd short everything to
everything.
He could just try to align it as best he can and hold it there to test. Glue
it if it works.

N

Franc made no mention of a zebra connection otherwise I wouldn't have
suggested epoxy. He did say that the original seemed to be attached
with some sort of adhesive so that's the reason for my suggestion.

On Mon, 15 Nov 2004 13:47:24 GMT, (Gerard Bok) put
finger to keyboard and composed:

On Mon, 15 Nov 2004 14:53:39 +1100, Franc Zabkar
wrote:

I've been given a Mercedes-Benz instrument cluster where a flex cable
has come adrift due to mishandling. The PCB is microcontroller based
and interfaces with an odometer via a flex cable. The other gauges
connect via long pins. AFAICS, the flex cable is attached with some
kind of glue - there are no clamps, no screws, no rubber strain
relief. The traces have a very fine pitch, and the cable is almost too
short to work on. The PCB is single sided, so I could probably drill
two holes

Are you very sure ?
'Very fine pitch' and 'single sided PCB' sound like a
contradiction to me. (Or as a euphemism for 'multi layer :-)

There are no thru-holes. All components are smt. The mechanic says an
instrument specialist told him that these flex cables are known to
separate in normal use, probably due to heat and vibration. He claims
these units are irreparable.

Here is a photo of the PCB:
http://members.optusnet.com.au/~fzabkar/pcb_trace.jpg

The connection point is above and between the two lamps. The traces go
directly to a microcontroller. Note that there are numerous test
points which look like thru-holes, but I believe these could be for
bed-of-nails testing (???).


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

On Tue, 16 Nov 2004 02:03:45 GMT, "NSM" put finger to
keyboard and composed:


"Ross Herbert" wrote in message
.. .

| I don't like your chances Franc. The only thing I an think of is
| electrically conductive epoxy.

If it's a zebra connection wouldn't that be worse? You'd short everything to
everything.

It's not a zebra elastomer connection. The copper traces of the flex
cable bear directly against the copper traces of the PCB. The two are
retained with some kind of glue. It's a very peculiar and IMO very
poor connection method. In fact, I've seen more adhesive on a postage
stamp.

He could just try to align it as best he can and hold it there to test. Glue
it if it works.

On closer inspection some of the printed copper traces on the flex
lead have delaminated. It appears this job is a lost cause. :-(

Anyway, here are some pics. The original 5MP photos have been
compressed to 90%, so the quality is poor.

http://members.optusnet.com.au/~fzabkar/label.jpg
http://members.optusnet.com.au/~fzabkar/instruments.jpg
http://members.optusnet.com.au/~fzabkar/flex_lead2.jpg
http://members.optusnet.com.au/~fzabkar/flex_lead.jpg
http://members.optusnet.com.au/~fzabkar/pcb_trace.jpg


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

"Franc Zabkar" wrote in message
...

| It's not a zebra elastomer connection. The copper traces of the flex
| cable bear directly against the copper traces of the PCB. The two are
| retained with some kind of glue. It's a very peculiar and IMO very
| poor connection method. In fact, I've seen more adhesive on a postage
| stamp.
| On closer inspection some of the printed copper traces on the flex
| lead have delaminated. It appears this job is a lost cause. :-(

I'd say so. I only see two choices, fix it, which may not be possible, or
replace it with something else. Depends on how much time and effort you want
to put into the task and how much value you will get for it.

You could start collecting the dead ones and use them as a garden border.

NM

There are no thru-holes. All components are smt. The mechanic says an
instrument specialist told him that these flex cables are known to
separate in normal use, probably due to heat and vibration. He claims
these units are irreparable.

Here is a photo of the PCB:
http://members.optusnet.com.au/~fzabkar/pcb_trace.jpg

The connection point is above and between the two lamps. The traces go
directly to a microcontroller. Note that there are numerous test
points which look like thru-holes, but I believe these could be for
bed-of-nails testing (???).

Yes, those "bumps" are for the bed of nail test points.

Looks like the ribbon and the pads on the circuit board are good, Why
not do a low-temp soldering?

Next time use a good older program like photo shop pro 5.x (ebay
hopefully) to crop it (this chops off the unneeded pixels) then shrink
it down to 50% to cut down bytes, 30% is minimum for smallest size
without losing too much details.

Cheers,

Wizard


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

On Wed, 17 Nov 2004 16:38:54 +1100 Franc Zabkar
wrote:

It's not a zebra elastomer connection. The copper traces of the flex
cable bear directly against the copper traces of the PCB. The two are
retained with some kind of glue. It's a very peculiar and IMO very
poor connection method. In fact, I've seen more adhesive on a postage
stamp.

I once had a basket case pocket calculator that someone else had taken
apart and destroyed. The damage they did consisted of breaking apart a
cable to board connection that was exactly like you are describing. I
was never able to come up with a repair method, but the one thing that
I thought might be worth trying was some sort of heat & pressure, like
layng a rag over it and then ironing it into place.

It's just possible that the adhesive is temperature sensitive.

I finally gave up and threw the calculator parts away, because I was
just never going to find the time to try to fix it.

-
-----------------------------------------------
Jim Adney
Madison, WI 53711 USA
-----------------------------------------------

On Thu, 18 Nov 2004 02:23:43 GMT, (Jason D.) put
finger to keyboard and composed:

There are no thru-holes. All components are smt. The mechanic says an
instrument specialist told him that these flex cables are known to
separate in normal use, probably due to heat and vibration. He claims
these units are irreparable.

Here is a photo of the PCB:
http://members.optusnet.com.au/~fzabkar/pcb_trace.jpg

The connection point is above and between the two lamps. The traces go
directly to a microcontroller. Note that there are numerous test
points which look like thru-holes, but I believe these could be for
bed-of-nails testing (???).

Yes, those "bumps" are for the bed of nail test points.

Looks like the ribbon and the pads on the circuit board are good, Why
not do a low-temp soldering?

My patience and eyesight aren't good enough for this. In any case an
email from a kind gent has cleared up the mystery for me. The
connection method is 3M Scotch 9703 conductive tape:

http://products3.3m.com/catalog/us/e...er/output_html

"Tape 9703 is an isotropic, electrically conductive, pressure
sensitive adhesive (PSA) transfer tape. It is a permanently tacky
system consisting of a PSA matrix with aligned conductive particles.
These particles allow electrical conduction through the adhesive
thickness ("Z-axis"), but not in the plane of the tape."

Next time use a good older program like photo shop pro 5.x (ebay
hopefully) to crop it (this chops off the unneeded pixels) then shrink
it down to 50% to cut down bytes, 30% is minimum for smallest size
without losing too much details.

Sorry for the abysmal quality. I should have made better use of Paint
Shop Pro, and I should have taken more care with the photography.

Here are the same photos with better processing:
http://members.optusnet.com.au/~fzabkar/pcb_trace.jpg (33KB)
http://members.optusnet.com.au/~fzabkar/flex_lead.jpg (19KB)
http://members.optusnet.com.au/~fzabkar/flex_lead2.jpg (17KB)
http://members.optusnet.com.au/~fzabkar/label.jpg (13KB)
http://members.optusnet.com.au/~fzabkar/instruments.jpg (13KB)

Cheers,

Wizard

I wonder if Scotch 9703 conductive tape may be a viable solution for
repairing broken flex leads?


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

"Franc Zabkar" wrote in message
...

| My patience and eyesight aren't good enough for this. In any case an
| email from a kind gent has cleared up the mystery for me. The
| connection method is 3M Scotch 9703 conductive tape:

Most interesting. I wonder if this is obtainable in service quantities?

N

"Jason D." wrote in message
...
There are no thru-holes. All components are smt. The mechanic says an
instrument specialist told him that these flex cables are known to
separate in normal use, probably due to heat and vibration. He claims
these units are irreparable.

Here is a photo of the PCB:
http://members.optusnet.com.au/~fzabkar/pcb_trace.jpg

The connection point is above and between the two lamps. The traces go
directly to a microcontroller. Note that there are numerous test
points which look like thru-holes, but I believe these could be for
bed-of-nails testing (???).

Yes, those "bumps" are for the bed of nail test points.

Looks like the ribbon and the pads on the circuit board are good, Why
not do a low-temp soldering?

Next time use a good older program like photo shop pro 5.x (ebay
hopefully) to crop it (this chops off the unneeded pixels) then shrink
it down to 50% to cut down bytes, 30% is minimum for smallest size
without losing too much details.



You don't even need that, you can get IrfanView for free, it'll resize,
convert formats, colors, etc not to mention view just about anything.

"NSM" wrote in message
news:Q7Xmd.96931$E93.47299@clgrps12...

"Franc Zabkar" wrote in message
...

| My patience and eyesight aren't good enough for this. In any case an
| email from a kind gent has cleared up the mystery for me. The
| connection method is 3M Scotch 9703 conductive tape:

Most interesting. I wonder if this is obtainable in service quantities?


Perhaps if you contact them they'd be willing to send you a sample of it?

On Thu, 18 Nov 2004 16:03:08 +1100, Franc Zabkar
wrote:

Here is a photo of the PCB:
http://members.optusnet.com.au/~fzabkar/pcb_trace.jpg

My patience and eyesight aren't good enough for this. In any case an
email from a kind gent has cleared up the mystery for me. The
connection method is 3M Scotch 9703 conductive tape:

http://products3.3m.com/catalog/us/e...er/output_html

"Tape 9703 is an isotropic, electrically conductive, pressure
sensitive adhesive (PSA) transfer tape. It is a permanently tacky
system consisting of a PSA matrix with aligned conductive particles.
These particles allow electrical conduction through the adhesive
thickness ("Z-axis"), but not in the plane of the tape."

In other words: self adhesive zebra strip :-)

--
Kind regards,
Gerard Bok

Here are the same photos with better processing:
http://members.optusnet.com.au/~fzabkar/pcb_trace.jpg (33KB)
http://members.optusnet.com.au/~fzabkar/flex_lead.jpg (19KB)
http://members.optusnet.com.au/~fzabkar/flex_lead2.jpg (17KB)
http://members.optusnet.com.au/~fzabkar/label.jpg (13KB)
http://members.optusnet.com.au/~fzabkar/instruments.jpg (13KB)

Cheers,

Wizard

I wonder if Scotch 9703 conductive tape may be a viable solution for
repairing broken flex leads?


- Franc Zabkar

That special tape will work! Key is clean both real good. Careful on
that flex cable, it is made of conductive paint printed on clear
plastic sheet. Many chemicals will attack that paint. This is same
kind of flex circuit with this printed on conductive paint.

Otherwise, there is room on PCB to drill two holes and use two thick
(say .050" to .100") strips of aluminum and screws & nuts, one thin
rubber strip from your junk stuff to pad the flex so both flex and
circuit board is squeezed together. Much better design IHMO!!

What type of display panel? LCD?

Cheers,

Wizard

On Thu, 18 Nov 2004 16:03:08 +1100 Franc Zabkar
wrote:

In any case an
email from a kind gent has cleared up the mystery for me. The
connection method is 3M Scotch 9703 conductive tape:

http://products3.3m.com/catalog/us/e...er/output_html

"Tape 9703 is an isotropic, electrically conductive, pressure
sensitive adhesive (PSA) transfer tape. It is a permanently tacky
system consisting of a PSA matrix with aligned conductive particles.
These particles allow electrical conduction through the adhesive
thickness ("Z-axis"), but not in the plane of the tape."

I agree with the others that this is a VERY interesting product, but I
don't think that it was what was used on the calculator I worked on
many years ago. That stuff did not look like tape, and it certainly
wasn't tacky any more.

There must be something else that is suitable for OE use. Perhaps it
is ultraviolet curing.

-
-----------------------------------------------
Jim Adney
Madison, WI 53711 USA
-----------------------------------------------

On Fri, 19 Nov 2004 00:27:34 GMT, (Jason D.) put
finger to keyboard and composed:

Here are the same photos with better processing:
http://members.optusnet.com.au/~fzabkar/pcb_trace.jpg (33KB)
http://members.optusnet.com.au/~fzabkar/flex_lead.jpg (19KB)
http://members.optusnet.com.au/~fzabkar/flex_lead2.jpg (17KB)
http://members.optusnet.com.au/~fzabkar/label.jpg (13KB)
http://members.optusnet.com.au/~fzabkar/instruments.jpg (13KB)

Cheers,

Wizard

I wonder if Scotch 9703 conductive tape may be a viable solution for
repairing broken flex leads?


- Franc Zabkar

That special tape will work! Key is clean both real good. Careful on
that flex cable, it is made of conductive paint printed on clear
plastic sheet. Many chemicals will attack that paint. This is same
kind of flex circuit with this printed on conductive paint.

Otherwise, there is room on PCB to drill two holes and use two thick
(say .050" to .100") strips of aluminum and screws & nuts, one thin
rubber strip from your junk stuff to pad the flex so both flex and
circuit board is squeezed together. Much better design IHMO!!

I had the same idea as you until closer examination of the flex lead
revealed that not all the conductive traces were intact. Maybe I can
shorten the lead, but it seems too short already. Of course this
presumes that I can obtain 3M tape 9703 in service quantities.

What type of display panel? LCD?

I don't know whether the display is an LCD because I've never seen it
working and I haven't removed it from the cluster. However, all the
other automotive displays I've seen have been vacuum fluorescent
types. In any case, I don't have the unit with me.


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

On Fri, 19 Nov 2004 00:27:34 GMT, (Jason D.) put
finger to keyboard and composed:

Here are the same photos with better processing:
http://members.optusnet.com.au/~fzabkar/pcb_trace.jpg (33KB)
http://members.optusnet.com.au/~fzabkar/flex_lead.jpg (19KB)
http://members.optusnet.com.au/~fzabkar/flex_lead2.jpg (17KB)
http://members.optusnet.com.au/~fzabkar/label.jpg (13KB)
http://members.optusnet.com.au/~fzabkar/instruments.jpg (13KB)

Cheers,

Wizard

I wonder if Scotch 9703 conductive tape may be a viable solution for
repairing broken flex leads?


- Franc Zabkar

That special tape will work! Key is clean both real good. Careful on
that flex cable, it is made of conductive paint printed on clear
plastic sheet. Many chemicals will attack that paint. This is same
kind of flex circuit with this printed on conductive paint.

Otherwise, there is room on PCB to drill two holes and use two thick
(say .050" to .100") strips of aluminum and screws & nuts, one thin
rubber strip from your junk stuff to pad the flex so both flex and
circuit board is squeezed together. Much better design IHMO!!

This is what 3M's 9703 datasheet has to say about "mechanical
clamping":

"To assure electrical resistance stability of tape 9703 in any
flexible circuit interconnection application, a mechanical clamp
should be considered in the design of the application. Without
clamping, over time, any shear stress or temperature excursions
applied to the bond line could result in an electrical open in the
bonded circuit. A well designed mechanical clamp will reduce the
environmental stress on the bond line and improve the electrical
reliability of the bond. Several types of mechanical clamps have been
used successfully including foam strips attached to lids or cases and
screw attached plastic clamps."


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

On Thu, 18 Nov 2004 22:01:40 -0600, Jim Adney put
finger to keyboard and composed:

On Thu, 18 Nov 2004 16:03:08 +1100 Franc Zabkar
wrote:

In any case an
email from a kind gent has cleared up the mystery for me. The
connection method is 3M Scotch 9703 conductive tape:

http://products3.3m.com/catalog/us/e...er/output_html

"Tape 9703 is an isotropic, electrically conductive, pressure
sensitive adhesive (PSA) transfer tape. It is a permanently tacky
system consisting of a PSA matrix with aligned conductive particles.
These particles allow electrical conduction through the adhesive
thickness ("Z-axis"), but not in the plane of the tape."

I agree with the others that this is a VERY interesting product, but I
don't think that it was what was used on the calculator I worked on
many years ago. That stuff did not look like tape, and it certainly
wasn't tacky any more.

There must be something else that is suitable for OE use. Perhaps it
is ultraviolet curing.

Here is another interesting product that bonds flex cable to coated
glass (eg LCD modules):

"3M™ Film 5552R electrically connects and mechanically bonds fine
pitch (less than 100 micron flexible printed circuit (Flex) -
especially copper/polyimide (PI) circuits - to Indium Tin Oxide (ITO)
coated glass substrates. Film 5552R is ideal for fine pitch systems
requiring high electrical conductivity and high reliability, along
with repairability. It offers low electrical interconnection
resistance and high stability and reliability over a wide range of
demanding environmental conditions."

And another one:

"3M™ Film 5460R electrically connects and mechanically bonds flexible
printed circuits (Flex) – especially copper/polyimide (PI) circuits –
to a wide variety of electronic substrates and metallizations,
including printed circuit boards (PCB) and flexible circuits. Film
5460R is ideal for high performance systems requiring high electrical
conductivity and high reliability, along with repairability."

And still another one (Z-Axis Adhesive Film 7303):

"Use to electrically connect and mechanically bond flexible printed
circuits to printed circuit boards and other flex circuits. Thermoset
adhesive film with anisotropic electrical conductivity through the
Z-axis. Designed for electrically and mechanically bonding polyester
and polyimide flex circuits to printed circuit boards, and other flex
circuits."


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.