Bongofish
November 22, 2017, 04:53:29 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
News: You can attach photos hosted by the forum rather than using an external image hosting site, this means they will stay forever and not disappear after a year or two.
 
   Home   Help Search Login Register  
Pages: [1] 2 3
  Print  
Author Topic: Attempt to repair LCD ribbon with 3M 9703 Tape  (Read 45004 times)
Aerendraca
Administrator
Hero Member
*****
Posts: 1068


View Profile
« on: January 26, 2013, 07:29:07 PM »

So on my build log I wrote about how impressed I was with the build quality of the AUO M170EG01 panel and that the ribbons looked good and strong, well they weren't! I pulled of one of the ribbons joining the LCD driver circuit to the screen with very little force, much disappointment ensued but I have another one so I will just have to be extra careful with that one.

So I thought to myself 'how can I make the best of this situation?' and decided that after dishing out a suggestion regarding using 3M 9703 tape to extend FFC/FPCs without having tried this stuff (oh and it can be used to extend FFCs, see post "Does DIY Beamer still exists?" in the Help section), perhaps I could use this opportunity to see if this tape can be used to fix LCD ribbons that have been accidently pulled off. Here goes my findings - Spoiler alert it didn't go as planned!  Shocked

Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1068


View Profile
« Reply #1 on: January 26, 2013, 07:54:40 PM »

So here's the offending ribbon cable:

* P1140534_979x734 01.jpg (52.21 KB. 979x734 - viewed 2739 times.)


* P1140539_979x734 02.jpg (62.16 KB. 979x734 - viewed 2162 times.)


And here's a closeup of the screens ribbon connector pads (x20 magnification) as viewed from the bottom; Thought the screen was the other way up when I took the pictures, Doh!

* P1140543_979x734 03.jpg (65.33 KB. 979x734 - viewed 2748 times.)


Now the right way up:

* P1140554_979x734 05.jpg (80.92 KB. 979x734 - viewed 2106 times.)


And here's the tape protected beneath the brown wax coated card:

* P1140548_979x734 04.jpg (92.59 KB. 979x734 - viewed 2273 times.)


« Last Edit: January 26, 2013, 07:58:11 PM by Aerendraca » Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1068


View Profile
« Reply #2 on: January 26, 2013, 08:03:48 PM »

So I cut a 2mm wide piece of the tape with a sharp knife.

* P1140555_979x734 06.jpg (76.07 KB. 979x734 - viewed 1851 times.)


Carefully placed it on the connector pads.

* P1140557_979x734 07.jpg (69.08 KB. 979x734 - viewed 1809 times.)


And peeled off the wax coated card part.

* P1140559_979x734 08.jpg (63.88 KB. 979x734 - viewed 2641 times.)


This is how it looked with the card removed.

* P1140564_979x734 09.jpg (61.56 KB. 979x734 - viewed 2230 times.)


Following this I painstakingly placed the FFC onto the tape, here's a magnified view:

* P1140565_979x734 10.jpg (71.25 KB. 979x734 - viewed 2519 times.)

The speckled strip you can see is the 9703 tape.

The closeup view.

* P1140567_979x734 11.jpg (67.6 KB. 979x734 - viewed 2144 times.)

Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1068


View Profile
« Reply #3 on: January 26, 2013, 08:08:06 PM »

I then cut some strips of that stuff you use to light bunsen burners (I forget what it's called), used it between a bulldog clip to clamp the FFC to the glass, and left it for 24hours as suggested by Chocochan.

* P1140568_979x734 12.jpg (77.08 KB. 979x734 - viewed 2184 times.)



* P1140571_979x734 13.jpg (80.98 KB. 979x734 - viewed 1714 times.)



* P1140570_979x734 14.jpg (51.42 KB. 979x734 - viewed 1878 times.)


---------------------------------------------------------------------------------------

24hours late I plugged the panel in, switched it on, and..................................











































Nope, didn't work! A big black bar on the screen where there should be an image. Sad

« Last Edit: January 26, 2013, 08:39:34 PM by Aerendraca » Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1068


View Profile
« Reply #4 on: January 26, 2013, 08:20:27 PM »

So I gently peeled of the ribbon again, and cleaned off the tape to start again from scratch this time taking some more time to make sure I was confident with each step before moving on.

After cleaning off the tape I noticed that the original 'glue' was still present and considered that this was preventing the 9703 tape from making a good contact with the pads.

Whilst looking through a microscope I carefully began to scrape off the old 'glue' until lovely (albeit slightly scratched) shiny pads were visible.

* P1150593_979x734 14.jpg (57.72 KB. 979x734 - viewed 1636 times.)


Cut a 1mm piece of tape (since I also wondered if I had too much waste last time), and once again gently placed it on the glass panel.

* P1150597_979x734 15.jpg (65.41 KB. 979x734 - viewed 1445 times.)


Here's a view through the microscope again.

* P1150602_979x734 16.jpg (72.33 KB. 979x734 - viewed 1640 times.)


If you compare this last image to the image from the last time I thought I has successfully stuck the FFC down, you will notice that you can see light through the gaps between the pads/copper traces. This means that this time I have successfully lined the FFC up with the corresponding connector pads!

Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1068


View Profile
« Reply #5 on: January 26, 2013, 08:38:47 PM »

Once again I plugged in the screen and switched it on.

Well this time I could see flicking multicoloured lines of pixels in the area which was previously black, but not a coherent image as expected. I was very disapointed with this and rechecked each pad through the microscope for continuity; Since you can see each piece of conductive material suspended in the tape with 20x magnification, it was possible to check to make sure that pads weren't accidentally touching by illuminating from below, and none were (I tripple checked!).  Huh   Huh   Huh

Well, it turned out that I had accidentally pierced a hole right through the FFC - I presume when I was scraping off the old 'glue' - which had killed about 5 lines to the panel.

* P1150602_979x734 16b.jpg (397.51 KB. 979x734 - viewed 2117 times.)


Not only that but on closer inspection I had also partially pulled away the other end of the FFC from the LCD driver circuit!!!! Arghhhhhh!!!!! Frustrating.  Angry

Conclusion:
Well although I failed to successfully fix the panel, I do feel that had I not pierced the FFC and ripped up the other end this fix would have worked. Since I have a few bits and pieces kicking around I might see if I can repair these issues and confirm this, but for now I think I will call this a fail.  Sad

« Last Edit: January 26, 2013, 08:52:18 PM by Aerendraca » Logged
bernard
Administrator
Hero Member
*****
Posts: 2584


pato mania


View Profile
« Reply #6 on: January 28, 2013, 06:07:33 AM »

 Embarrassed Tongue  Cry

mixed feelings from here too. congrats for the surgeon work.
Logged
Chocochan
Full Member
***
Posts: 82


View Profile
« Reply #7 on: January 28, 2013, 01:40:50 PM »

Hey Aerendraca.

Sorry to hear your repair didn't work out. But something has been nagging me in your picture of the cut piece 3M 9703 tape. You cut a 2mm wide strip right? From the way it looks, to me it looks like it's in the wrong direction. What I mean to say is, the direction of conductivity runs up and down along the WIDTH of the tape, not the LENGTH. You need to cut a piece the width of the cable, then cut it down to fit the space. I'm sure you just used the wrong axis as there shouldn't be wax tape on the ends, only on the top or bottom (unless you did that, then I have no clue what went wrong).

There may still be a chance to fix it if you can expose the super thin contacts within the damaged cable and add a new bit of a FFC on the end act as an extension and create new contacts. Another idea is perhaps thin strips of copper tape? I wish I was there to help you! Sad

EDIT: A tip I learned is if you gently heat the tape when it's attached with the tip of a soldering iron, you can soften it enough to gently loose it off. Then use isopropyl alcohol (rubbing alcohol to those in the US) to clean up the residue along with gentle scraping with a blade.
« Last Edit: January 28, 2013, 01:44:40 PM by Chocochan » Logged
bernard
Administrator
Hero Member
*****
Posts: 2584


pato mania


View Profile
« Reply #8 on: January 28, 2013, 04:55:13 PM »

He mentioned he saw the contacts through the light through the microscope.  I would guess it is impossible that he couldn't have noticed traces going perpendicular(?)  Shocked
Logged
Chocochan
Full Member
***
Posts: 82


View Profile
« Reply #9 on: January 28, 2013, 05:17:19 PM »

He mentioned he saw the contacts through the light through the microscope.  I would guess it is impossible that he couldn't have noticed traces going perpendicular(?)  Shocked

The tape is made of a pressure sensitive acrylic adhesive filled with conductive particles. I don't think you would see any traces in the tape itself, but the particles are spaced out enough to be insulating from left to right, but conduct only top and bottom. This is mentioned in the PDF I linked to in "Does DIY Beamer still exist?"

Here's a quote from the PDF:

"3M™ Electrically Conductive Adhesive Transfer Tape 9703 is a pressure sensitive adhesive (PSA) transfer tape with
anisotropic electrical conductivity. The PSA matrix is filled with conductive particles which allow interconnection between
substrates through the adhesive thickness (the “Z-axis”) but are spaced far enough apart for the product to be electrically
insulating in the plane of the adhesive. The PSA tack properties and lack of any thermal curing make tape 9703 easy to use
in assembly operations.

Tape 9703 electrically connects and mechanically bonds medium pitch flexible circuits with other flexible circuits (flex),
rigid printed circuit boards (PCB) or LCD screens. Electrically conductive tape 9703 offers good adhesion to common
PCB substrates such as copper, gold, FR-4 epoxy, Kapton™ polyimide and polyester films. Stable electrical performance
in any flexible circuit interconnection application may require mechanical reinforcement (clamping).

Tape 9703 also electrically connects and mechanically bonds EMI/RFI shield and gaskets to metal frames and enclosures.
The low contact resistance and tape construction result in good EMI performance. Tape 9703 can be applied as die cut
parts or in roll form and has good adhesion to common EMI/RFI substrates such as aluminum, stainless steel, and smooth
gasket materials."

It mentions mechanical reinforcement may be required to ensure stable electrical performance in flexible circuits. Perhaps something on the join would've helped? I still stand by my first guess. The tape was used in the wrong axis, either that or the pitch was too small for the tape (it's supposed to work down to 0.4mm pitch)?

Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1068


View Profile
« Reply #10 on: January 28, 2013, 05:40:46 PM »

Well here's what I found when I analysed the tape:

The tape that I have has the 'lines' of conductivity running down the length of the tape and not the width, let me show you:

So here's the piece I have cut off the main bulk of the tape (ignore that I've chopped a bit from the right side), oriented so that the length goes from the top of the screen to the bottom:

* P1180719smallPaint.jpg (142.97 KB. 816x612 - viewed 1481 times.)


Here's that same piece flipped over horizontally:

* P1180718smallPaint.jpg (133.79 KB. 816x612 - viewed 1233 times.)


All I've done here is fold back the white waxy paper overhang and weighed it down with a scalpel:

* P1180717smallPaint.jpg (144.37 KB. 816x612 - viewed 1226 times.)


And here it is again with 20x magnification. The orientation is the same as the previous images.

* P1180716smallPaint.jpg (142.7 KB. 816x612 - viewed 1791 times.)


You can see here that the distribution of 'specks' is focussed top to bottom or down the length of the tape.

Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1068


View Profile
« Reply #11 on: January 28, 2013, 05:47:41 PM »

Since I was playing with this stuff anyway I thought I'd go further zooming in on this stuff. Note due to the way the tape was applied to a microscope slide Left to Right is along the length of the tape.

30x mag:

* P1180714smallPaint.jpg (146.78 KB. 816x612 - viewed 1183 times.)


40x mag:

* P1180708smallPaint.jpg (124.75 KB. 816x612 - viewed 1116 times.)


100x mag:

* P1180710smallPaint.jpg (87.08 KB. 816x612 - viewed 1091 times.)

The bubbles you can see are air that has been trapped between the coverslip and the tape.

400x mag:

* P1180711smallPaint.jpg (61.08 KB. 816x612 - viewed 1285 times.)

Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1068


View Profile
« Reply #12 on: January 28, 2013, 06:01:14 PM »

Quote
It mentions mechanical reinforcement may be required to ensure stable electrical performance in flexible circuits. Perhaps something on the join would've helped? I still stand by my first guess. The tape was used in the wrong axis, either that or the pitch was too small for the tape (it's supposed to work down to 0.4mm pitch)?

Mechanical reinforcement way well be a good point, although since there is no space for any kind of clamping device to be fitted along the edge of the glass there's not much I can do about that.

As for the orientation of the tape, I believe that I had the tape oriented in the right axis, although to be honest this is not so much of an issue as I checked to make sure that the 'specks' weren't connecting between tabs under a scope, as far as I could tell no cross connections.

I believe the biggest problems were the hole I punched in the ribbon (such a dunce), and the partial pull off of the other end of the ribbon, see images below:

Highlighting the rip in the ribbon.

* P1150605smallPaint.jpg (125.73 KB. 816x612 - viewed 1422 times.)



* P1150597smallPaint.jpg (153.07 KB. 778x531 - viewed 3785 times.)

Logged
Chocochan
Full Member
***
Posts: 82


View Profile
« Reply #13 on: January 28, 2013, 07:48:58 PM »

Well, then maybe your problem is the pitch then. This is a scan of part of the instructions I received with the tape:


* tape.jpg (16.26 KB. 640x384 - viewed 1365 times.)


This is also the same way the guy in the YouTube video described how the tape carries signals. I just followed this and it worked flawlessly. I'm going to remove it soon as the extension is too long, but here's a pic of it working before I do that (I'm going to use a different screen anyway). The join is covered with white electrical tape and if you fiddle with it, it does flicker, but that is to be expected. Once installed in a case it won't ever move!


* SANY0050.JPG (294.65 KB. 1280x720 - viewed 1574 times.)


I still think you can save that cable, but it'll require steady hands and a bit of effort. I'll have to look for a similar one and see what I can do (I'm feeling optimistic right now which is a first). Really, I only used enough tape to cover the exposed contacts and it worked (just to note, cable is 45 pin 0.5mm pitch).

« Last Edit: January 28, 2013, 08:20:53 PM by Chocochan » Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1068


View Profile
« Reply #14 on: January 28, 2013, 08:55:54 PM »

I never really expected this method to work since I think the pitch is smaller than 0.4mm however since I have seen this stuff under a scope I figure it can be used on smaller pitched ffc but there becomes an element of chance as to whether it will work, this is based on an assumption that 3M would guarantee this tape to work down to pitches 100% of the time where probability of success reduces proportionally to decreasing pitch.

Anyway, I believe too that this ffc could be fixed, so here's how I went about trying to fix the hole and the ripped up part of the ffc:

The ribbons which join the driver circuit to the glass are actually joined in pairs at the driver board circuit, so I had to carefully cut the ffc at a non-crital point and then gently peel the broken ffc away. Here's how it looked:

* mini-P1170606 a01.jpg (87.03 KB. 800x600 - viewed 1079 times.)


The broken ribbon. As you can see the hole I made ripped a section out when I peeled the ribbon away from the glass.

* mini-P1170610 a02.jpg (65.55 KB. 800x600 - viewed 1318 times.)


The other end of the ribbon. You can't see it very well here but where there is the cut section, part of one trace got cut whilst peeling the ffc from the circuit board. This meant that I would need to chop some of it off, but that happens later.

* mini-P1170611 a03.jpg (60.89 KB. 800x600 - viewed 1121 times.)


Back to the glass end of the ffc, and here you can see how I chopped a section off to bring it back to where I punched the hole.

* mini-P1170619 a04.jpg (78.24 KB. 800x600 - viewed 1880 times.)


After cutting it back I taped it down to my cutting mat and began gently sanding away the plastic coating with wirewool.

* mini-P1170621 a05.jpg (107.93 KB. 800x600 - viewed 1352 times.)


That wasn't working too well so I decided to use one eye in the microsope and a scalpel to scrape the plastic coating away.

* mini-P1170638 a06.jpg (84.35 KB. 800x600 - viewed 1234 times.)

Logged
Pages: [1] 2 3
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Valid XHTML 1.0! Valid CSS!