LCD image reversed Line6 Helix

[Audiorepair]

What could possibly cause an LCD to suddenly reverse its image?

I have a Line6 Helix here because of this. 
When I first booted it the boot Splash screen was normal, then after booting it was reversed as in the photo.
Now it is always reversed.

There is one other instance of this you can find online but no solution.

I feel it has to be firmware, though the customer contacted Line6 and they suggested not, and that it was a hardware or ribbon cable issue.

I did read an online comment suggesting along the lines that maybe it has somehow got itself into a language mode that writes right to left (Arabic e.g.)
This seems to me to be the likeliest suggestion, unless there is some other way this could possibly happen.

Interesting nonetheless.

[coromonadalix]

must be HW issues  bad connection

some did complain about this and fw updates or downgrades did not recover ...    ribbon, bad ribbon insertion ...     

you have to dismantle it to check  the lcd markings if any and get datasheets,  could be control lines  etc ...


https://line6.com/support/topic/24355-flipped-lcd-screen/

unless you're one of thoses who complained on other forums ?

[Audiorepair]

No I haven't been on any forums about this.

So is it possible to do this via a bad ribbon cable or messed up control lines?

My head is having a hard time believing this.

[Audiorepair]

Or another way of putting it is, could you add a hardware function to a LCD that reversed the image like this?

[Nominal Animal]

It is possible for the LCD controller to have one to three hardware pins to set the display orientation.  This would explain why the manufacturer said it is a ribbon cable issue.  (Perhaps a Return of the Tin Whiskers, what with lead-free solder and dense connections?)

Most matrix LCD and TFT controllers I've seen –– I'm only a hobbyist who interfaces small displays in the 1" - 5" range to Linux-based appliances like routers and such for non-technical users to see the state, without risking changing the settings –– do have commands for rotating and mirroring the display.

If only there was more information about the LCD unit, and what controller it might use!

The display modules I use have a controller chip embedded in the display or flat flex cable.  See for example this 4.3" IPS one from BuyDisplay/EastRising.  The chip is on the display module itself, underneath the edge of the display.  It directly controls all the hundreds of LCD matrix pins.  This particular display uses ST7262E43 controller/driver (datasheet PDF).  Although this display (datasheet PDF) does not have orientation pins exposed on the 40-pin ribbon cable, the ST726E43 controller does have exactly such mirroring pins, HDIR and VDIR, on pads 157-161 (two consecutive pads each).  I believe most controller/drivers do have such pins, although I haven't verified it.  In any case, it's up to the display module manufacturer to decide whether they're exposed on the FFC cable or not.

Without knowing the FFC pinout, or the exact controller used on your display module, there isn't much anyone can do to help.  The display modules are rarely custom jobs – although there are a lot of variants on the FFC pinouts; I do believe it is not at all expensive to get a run with custom pinout – so looking for any kind of model number information on the display module and its Flat Flex Cable might lead to manufacturer and documentation.

[coromonadalix]

some lcd have control lines to do mirroring or image inversion, it could be a simple bad contact

speculations until  you get the lcd datasheet if any, maybe there is a control pin for that,  or even if some  did say the Fw was not at cause ... 

we do see the image, all seems perfect,  but inverted ...




as an example: we saw here some meter(s) oled substitution projects,   there was code and other was simply hardware control lines ..

[tom66]

Are other functions of the LCD unavailable for instance brightness/contrast control, I wonder if a broken I2C bus (commonly used for display setup) might allow the display to operate with an image but configuration would be wrong.

[Audiorepair]

Photo of LCD markings.

[Audiorepair]

Nothing on the underside.

[Audiorepair]

Goes to this daughter board, so one cable.

[Nominal Animal]

Photo of LCD markings.

The display module is a 6.2" 800x480 HannStar HSD062IDW1-A00 (also known by 721CR60012-A0, apparently relatively common in automotive applications) with a 60-pin FFC.  Good news: you can get a replacement one from Aliexpress and eBay for about 20€.  All HSD062IDW1-Axx models should be interchangeable by what my searches indicate, but do check carefully for yourself; I'm often wrong.  (Don't let my writing style fool you into thinking I'm a pro; I'm just an uncle bumblefug hobbyist.)

Even better, I located a datasheet.  The problem is pin 35 in the FFC, SHLR.  It is right next to pin 36 (digital supply).  I'm willing to bet pin 35 should be grounded, but somehow pin 36 is making a tiny contact to it.  Pin 1 and pin 60 should be marked on the FFC, but if not, the RGB data is on pins 4-27; i.e. the end with many parallel data lines has pin 1.

What I would do, is disconnect the display (when the device is disconnected from mains and not powered on!), and check the continuity between pin 35 and 36 with a digital multimeter in Ohms/resistance mode; and between pin 35 and pin 38 (digital ground).  Pin 37 is the UPDN, i.e. vertical mirroring.

If you are very, very lucky, it is just slightly shifted FFC within the connector, and reinserting the FFC will fix the image.  Or that you can see some discoloration or suspiciously sharp-looking tinning on pads 35 and 36, and some light, very careful scraping with a scalpel can fix it.  Otherwise, the fix is a bit difficult, as you need to find where the unwanted connection is –– or just replace the display.  I am hoping it is in the connector end of the FFC cable, and not in the display side.

[Audiorepair]

Well there you go.
Oh me of little faith.

[Nominal Animal]

Fixed it by re-seating the FFC in the connector?  So pin 35 is no longer contacting pin 36?

[coromonadalix]

Fixed it by re-seating the FFC in the connector?  So pin 35 is no longer contacting pin 36?

could be ...   maybe there was some pins shift in the socket,  theses fpc are very narrow, 0.5" pitch i think,  even a bad insertion could deform  the flex pcb traces etc ...  bad connection,  or badly clamped socket holder / snap / lock


time will tell

[Nominal Animal]

The FFC is 60-pin, 0.5mm (0.019685") pitch, so very, very dense at the connector, which has individual contact pins/springs.  At the middle of the FFC, the distortions are greatest (the connector slot is very precisely the same width as the FFC), so it is no wonder pins 35 and 36 (of 60) can sometimes overlap a tiny bit – even from say thermal cycling, I'd guess.  The contact itself can be minuscule, since pin 35 is just a logic level high impedance input pin, and pin 36 is one of the digital supply voltage pins.  Depending on the contact materials (ENIG, or just lead-free tinned), tin whisker is an actual possibility here too.

It would be very nice if OP would inform others having the same issue how they themselves solved it and how it may occur...  You know, pass the help forwards.

[Audiorepair]

I just blew on and reseated the 3 ribbon connections and reassembled.

Then it worked.  I'll check tomorrow if it is still working well.

Thanks for all the help and info, I'm sure someone else with this problem will find this thread now.


I do just have one query though, if this was caused by 2 adjacent ribbon lines shorting, would this happen by manipulating the ribbons and trying to recreate this short after booting, or is this line just polled on startup and ignored subsequently?
It seemed a bit odd that when I first powered up, the boot display was the right way round, but as soon as it got to the main program screen it flipped and stayed flipped.


Cheers.

[Nominal Animal]

Thanks for all the help and info, I'm sure someone else with this problem will find this thread now.

At minimum, post a short report at each other forum/thread you posted to about this.  Do not assume web search engines will find the one thread where a solution was discovered – especially since you didn't even initially post that it got fixed, just posted an image showing that it is now fixed.  No web search engine is able to tell that this particular thread contains fixes.

To be blunt, you owe it to those who helped you, to pass it forwards.

if this was caused by 2 adjacent ribbon lines shorting, would this happen by manipulating the ribbons and trying to recreate this short after booting, or is this line just polled on startup and ignored subsequently?

It is typically sampled at the start of every frame, once every display refresh, so about 60 times a second.

It is absolutely possible to recreate the mirroring by connecting pins 35 and 36 using a suitable resistor.  (A strong direct short, say using a screwdriver to poke at the gap between those pins on the daughter card, may create a direct short between the digital VCC and GND, and damage something.)  It's just that at 0.5mm pitch, it's beyond my own finger dexterity; if I were to try and do that, I bet something would go Pop! and release the magic smoke...

[Audiorepair]

Thanks for all the help and info, I'm sure someone else with this problem will find this thread now.

At minimum, post a short report at each other forum/thread you posted to about this.  Do not assume web search engines will find the one thread where a solution was discovered – especially since you didn't even initially post that it got fixed, just posted an image showing that it is now fixed.  No web search engine is able to tell that this particular thread contains fixes.

To be blunt, you owe it to those who helped you, to pass it forwards.

To be blunt, I was at the end of a long day and drove home and ate my dinner.
Then I posted again.

[Nominal Animal]

To be blunt, I was at the end of a long day and drove home and ate my dinner.
Then I posted again.

I did not mean it as an admonishment/criticism/negative, or to annoy you, but more as an observation and general suggestion on maximizing the usefulness of such solutions – including to future oneself.  I am blunt and direct, and have no social skills to speak of.  I may appear as someone very knowledgeable, but in fact, I only have good search/reading skills, and have to spend a lot of effort getting my output in even a roughly acceptable state.  Very often, I fail to consider the social subtext. 

For me, it is extremely important that solutions found are reported and made available to others suffering from the same problem.  This is what makes it worth the effort of research and answering: it must help more than just one person asking.  The best case, in my opinion, is when the help-ee posts a summary describing the issue and the solution as they understand it (a quick single paragraph), for maximum search engine discoverability; and in turn helps someone else later on.

If you want to know the basis and reasoning of this, see the why writing style and grammar matters in posts in the Beginner section.

You could say this is my small way of making the world a better place.  A tiny way, microscopic, sure; but it is still better than nothing.  All that is necessary for evil to prevail, is for good people to do nothing.

[coromonadalix]

@Nominal  not to start an debate

While i understand you're statements, the OP is not under any obligation(s) to post elsewhere to provide a solution, for now if it work ... 

YOU can write to other forums ...  because in a sense  you'll be responsible in a way of the posted threads ...

And if you do google search(s)   eevblog  will come out as an answer ..

[Nominal Animal]

OP is not under any obligation(s)

Technically true, but even plain Game Theory says it is beneficial of OP themselves in the long run to do so for every forum they posted the question about this.  Enlightened self-interest and all.

[Audiorepair]

I have already stated here, that this is the only forum I have posted to on this subject.

All the relevent info required and the solution to this issue is now easily found on Google, including the part number of the display.


My work is done here, my customer's Helix is repaired, anyone with the same problem will easily find out what to do.

[Audiorepair]

Thanks for all the help and info, I'm sure someone else with this problem will find this thread now.

if this was caused by 2 adjacent ribbon lines shorting, would this happen by manipulating the ribbons and trying to recreate this short after booting, or is this line just polled on startup and ignored subsequently?

It is typically sampled at the start of every frame, once every display refresh, so about 60 times a second.

It is absolutely possible to recreate the mirroring by connecting pins 35 and 36 using a suitable resistor.  (A strong direct short, say using a screwdriver to poke at the gap between those pins on the daughter card, may create a direct short between the digital VCC and GND, and damage something.)  It's just that at 0.5mm pitch, it's beyond my own finger dexterity; if I were to try and do that, I bet something would go Pop! and release the magic smoke...

OK so now I am doing this in my own time that the customer is not paying for,  I query your assertion above.

I looked at the manual for this display, and on page 19 it gives the power up/power down sequence.
The pin35 you correctly identified as the pin that will flip the display, is designated SHLR.

However it seems, as I postulated, that this pin is polled at Startup as part of the Initialisation process, it is not constantly monitored subsequently.
So no amount of physical stimulation of the ribbon cables after boot up will be able to flip the display, or indeed poking about with a screwdriver as you imply.

This from the data sheet:
Notes: Data include R0~R7, G0~G7, B0~B7, HSD, VSD, DCLK, SHLR, UPDN, DE
MODE, RSTB, STBYB, SHLR, UPDN, DITH

http://static1.1.sqspcdn.com/static/f/489821/26342048/1435254000047/HSD062IDW1-A00.pdf?token=wedVmeycvRpgeAo4dr1NFba11vw%3D



Sorry for the crap image, I couldn't easily find a way to capture the particular PDF page and save it on my home Mac as I do with my Work PC, so I just took a photo on my phone.
I guess I'm used to making the best use of my time.

[Nominal Animal]

It is typically sampled at the start of every frame, once every display refresh, so about 60 times a second.

I looked at the manual for this display, and on page 19 it gives the power up/power down sequence.
The pin35 you correctly identified as the pin that will flip the display, is designated SHLR.

However it seems, as I postulated, that this pin is polled at Startup as part of the Initialisation process, it is not constantly monitored subsequently.

HannStar does not specify which controller it uses, and the datasheet isn't exactly clear.  In the image 6.5, SHLR and UPDN are not part of the startup or shutdown sequence, but are grouped with "data include R0~R7, G0~G7, B0~B7, HSD, VSD, DCLK, SHLR, UPDN, DE [...] MODE, RSTB, STBYB, DITH".  This implies they are continuously sampled, not sampled only at power-up.

The controllers I have used, do sample the polarity/mirroring pins for every frame, including the ST7262E43 I already mentioned.  Some, like ST7789V2, do not have such input pins, but do have a command to select orientation and mirroring of the display.  The displays like ER-TFT070IPS-4 with ST7277 controllers (that do have such an input pin) do not export that pin to the FFC.  These are for 800×480 displays, like the one you have, by the way; I mostly use smaller ones, 320x240 and 480x320 ones.

Perhaps I should have worded that differently, say "in my experience,", instead of "typically", but if you look at display modules with datasheets freely available (ILI9341, ILI9488, ILI9881, ST7701, ST7789, NV3052, NT35510), they definitely allow changing the orientation at runtime via a dedicated command.  Perhaps it is an error for me to make any inferences from these to all display modules.

Simply put, in my experience across the various controllers, the direction is sampled at every new frame, and can be switched without turning the display off.  It is quite rare for anything to be only sampled by the display controller at power-on, and ignored afterwards: none of the displays I have do that.

So no amount of physical stimulation of the ribbon cables after boot up will be able to flip the display, or indeed poking about with a screwdriver as you imply.

Are you sure?  Did you verify this in practice, or is this just how you understand the datasheet?  In my opinion, the datasheet isn't clear on this.  In particular, R0~R7, G0~G7, B0~B7, HSD, VSD, DCLK and DE are the digital video signals, and they definitely aren't simply sampled at power-up; they are continuously sampled, as otherwise the display would be all in single color.

I am not saying it is impossible, I am only saying it would be different to all the display controllers I have.  Granted, I don't have that many, maybe a two dozen, but I do have considered and compared the most commonly available openly documented ones, and read their datasheets and pinouts, and examined the open source libraries used to drive these from microcontrollers (which is basically what I do for a hobby, as I've explained above).  I myself have plenty of time for my hobby, but only a small budget.

[Audiorepair]

Ok,  I don't see why any designer would want to include the possibility of LCD flip where such a possibility would never be desirable.

They would surely just wire this out and use the INIT stage to set it as required.


Look, this is pretty much a moot point.
The problem here is clearly to do with dirty ribbon cables, reseating them is the solution.
Problem solved.


There is little point in persuing this any further IMHO.