FORD technical service bulletin : ICC touch screen display - Page 5

XRPete · 10-10-2019, 07:44 PM

Quote:

Originally Posted by JasonACT

I'm afraid we're going to have to stop testing until that MOSFET is replaced...

image

I've disconnected my gate input, MOSFETs are basically driven by a capacitor that you charge via the gate - power in that cap switches the device on.

What I've found through testing while I have no gate input connected is, the screen doesn't turn on. The large and small CPU do. Everything works except the screen. I can't even see lines or a faint image using a torch.

Then I charge the MOSFETs cap by connecting it and then disconnect the gate again.. the cap remains charged (as they do) and the screen backlight turns on and stays on... until the cap finally discharges after a lot of time. Then the screen turns off and remains off. Everything else continues to work - recharge the cap again - screen turns back on and it's displaying the same as it should.

This is interesting for sure....

ill see if can find another mosfet from Jaycar or a local eletronics repairer.

I did call a local repairer today who said that can inspect the board from $85 for the first hour, and advise the status of the board after that. but the felt that dealing with a component like this could be big money wasted.

I also called Fords Parts department, and discussed the current dilemma with this icc and they advised that they havent received and kind of CSI or service bulletin etc that would deem this component faulty or classed for recall. they also told me that they do sell screens new for $1300, and have been offering second hand as an option for $700. but would guarantee me an solution as to where they would swap the unit or charge the replacement until i bring my car in.

I didnt get a chance to call SWS Australia today, however the local repair did advise that calling them would probably lead me no where, as they would probably redirect me back to ford.

this sounds similar to situations Ive experience as well at work, as i work for a mining parts/re-manufacturing department too.

JasonACT · 10-10-2019, 07:51 PM

The datasheet states "Drain connected to the tab".

If you are going to try a different MOSFET - just remember the heatsink (tab) is the drain, so that's why the middle pin is shorter and unconnected. That's where the +ve power is coming in from.

JasonACT · 11-10-2019, 08:48 AM

Quote:

Originally Posted by XRPete

Ok, so i use an new grounding point in the car. and the 3 pin mosfet which i thought i blew, is reading at I (on) key position 11.89v @ top pin and 21.70v on the bottom give or take some mv's.

Some thoughts:
1/ I had just assumed you got things mixed up here, because while I got those voltages, my readings are reversed - 20v top pin, 11v bottom.
2/ Since it has popped it's possible the readings will be weird.
3/ That MOSFET is rated for +/-20v gate voltage (like many are).
4/ Mine only just exceeds the 20v maximum, so my repeated bench testing might not ever show the problem of the LCD dying.
5/ Yours is going well over! Typically, 10v is used in circuits, except for logic level MOSFETS that work on 3.3v-5.0v - so 20v is the max that should safely be passed into them.
6/ I think that MOSFET dims the display at night time, it's going to get hotter at that time, so it needs to be cooled.
7/ Jaycar don't have MOSFETs in the size you need, but there's a couple there which are larger (with the same +/-20v gate specs) which may work, except night time dimming will be an "unknown" because the board has been calibrated to work with a different one. I'd be looking at the IRF1405 - it has the same pinouts and I'd somehow add a small flat heatsink.
(That's enough thoughts.)

Citroënbender · 11-10-2019, 10:12 AM

Re the failed semi: I use RS and Element 14 when Jaycar don’t carry an item. RS are sort-of convenient because I can “click and collect” from their warehouse/office with nil handling fee.

XRPete · 11-10-2019, 10:26 AM

Quote:

Originally Posted by Citroënbender

Re the failed semi: I use RS and Element 14 when Jaycar don’t carry an item. RS are sort-of convenient because I can “click and collect” from their warehouse/office with nil handling fee.

Thank you for that Jason.
I was just about to respond with the same information.
I actually had my brother in law inspect the board last night, he’s a gm for jaycar. It was just luck that he was down and could look at it for me.

He advised to go through element14 as jaycar only stock a 60v 16a through hole type which probably wouldn’t work, he was also saying that damaging a mosfet can give reverse reading, rare but can happen apparently.

Inspecting both boards, he couldn’t see any components blown or traces broken and he’s saying to start by replacing this mosfet too.

I’ve called a local repairer to see if they can procure this part, not I’ll try and order it from element14 or alibaba express as they seem to have some.

Also arranging a new rework/solder station today that will can handle soldering work of this nature.

happy1 · 11-10-2019, 10:37 AM

I get my parts from:

https://au.mouser.com/

and

https://www.x-on.com.au/

The best search engine is often on the Mouser site, but the companies work together, and after finding the part number I order from the X-on site as they are located in Australia.

PS: Mouser and X-on also have an impressive range of electrical plugs / connectors.

XRPete · 11-10-2019, 11:28 AM

Ok, I just got off the phone with this electronics repairer and said that this is a cheap Chinese mosfet and to order MTP15N06 (M stands for Motorola) or STP15N06 (STMicroelectronics) which are genuine manufacturers, best to get from eBay.

He then asked me what is this out of and asked me if I’ve tested the mosfet, which I did and He said, that it sounds like my mosfet seems fine but I could have blown the diode that built into it.

He then went on to say, he knows the problem and stated that the board needs recoding by for, that ford have failed to tell there customers that when changing a battery to keep constant power to the vehicle and that once’s the board on these ICCs are drained of power after a certain period of time, loss there codes and need recoding.

I’m finding the information he gave me questionable, as my screen died without reason. I didn’t change battery or have a flat. I just got in the car and started it and the screen was dead.

raceteam · 11-10-2019, 11:44 AM

That information seems inaccurate. I've seen an FDIM disconnected for weeks. If that was the issue, then instead of replacing a failed unit, Ford would plug the IDS in and reactivate it.

JasonACT · 11-10-2019, 11:56 AM

I googled those 2 parts, the pictures show them to be the same size as the Jaycar ones. My work mate who's screen died was also like yours - didn't change the battery, didn't jump start the car and didn't have a flat battery. It just died one day.

XRPete · 11-10-2019, 12:01 PM

Quote:

Originally Posted by raceteam

That information seems inaccurate. I've seen an FDIM disconnected for weeks. If that was the issue, then instead of replacing a failed unit, Ford would plug the IDS in and reactivate it.

I just got a response back from someone I contacted on Facebook, who does reprogramming for icc and keys in Canberra. He’s saying this issue is due to low power fault in the car and then there is a spike which fry’s the screen or goes blank.

See response below.

http://www.fordforums.com.au/photos/...80DF93BA2.jpeg

JasonACT · 11-10-2019, 12:24 PM

The ACM (radio/amp part) needs recoding. Forscan will do it, but you need an OBDII device with the high-speed and medium-speed switch - specific for Fords. I got one from ebay quite cheaply. Forscan needs a license too, but you can get a two month free trial pretty easily.

Initially, I just disassembled the ICC I had bought and swapped the LCD over - no problems with my own radio and the new screen. Then I put the ICC with my old screen and the new ACM in.. "Radio Code Error" showed up and I used Forscan to "recalibrate" it. Fixed.

Every time you reconnect power to the unit you will get the error while on accessories. You need to turn the key to ignition and the code error goes away once Forscan has programmed it correctly. It remains properly paired until it's unplugged from the battery again.

I'm skeptical it's not fixable. For starters, you've connected up a 9v battery and got the screen to light up. That's not the screen which is broken, it's the circuit powering it which is what we are trying to repair here.

hellrc5 · 11-10-2019, 01:23 PM

Really strange that the mosfet on one of those pics is really misaligned while it is really straight on the other panel as it should be. Misaligned components on mass produced boards assembled by robots are really rare especially if i consider the size of the component, even if it was replaced by hand it should be one of the easiest parts to replace. I know as long as it is connected to the PCB and the bottom too (cooling) it does not make any difference but to me it looks likely that this component has already been disturbed by someone.

Sent from my Note7 FE - Flames Extinguished edition

XRPete · 11-10-2019, 01:23 PM

Quote:

Originally Posted by JasonACT

The ACM (radio/amp part) needs recoding. Forscan will do it, but you need an OBDII device with the high-speed and medium-speed switch - specific for Fords. I got one from ebay quite cheaply. Forscan needs a license too, but you can get a two month free trial pretty easily.

Initially, I just disassembled the ICC I had bought and swapped the LCD over - no problems with my own radio and the new screen. Then I put the ICC with my old screen and the new ACM in.. "Radio Code Error" showed up and I used Forscan to "recalibrate" it. Fixed.

Every time you reconnect power to the unit you will get the error while on accessories. You need to turn the key to ignition and the code error goes away once Forscan has programmed it correctly. It remains properly paired until it's unplugged from the battery again.

I'm skeptical it's not fixable. For starters, you've connected up a 9v battery and got the screen to light up. That's not the screen which is broken, it's the circuit powering it which is what we are trying to repair here.

The HS-MS Can-Bus that you have is that a elm327 modified?
I was thinking of purchasing one of ebay today.

The screen for sure isn’t broken, as ti did the backlight test with the 9v battery and it worked fine for me to. it seems when he said screen, he is meaning the screen unit (Fdim) including the boards we are discussing.

JasonACT · 11-10-2019, 01:34 PM

Yep, modified ELM327:

https://www.ebay.com.au/itm/1Pcs-For...53.m2749.l2649

(Says Bluetooth - but I'm using the USB cable - didn't even think to try it on my Bluetooth tablet.)

XRPete · 11-10-2019, 03:41 PM

Quote:

Originally Posted by JasonACT

Yep, modified ELM327:

https://www.ebay.com.au/itm/1Pcs-For...53.m2749.l2649

(Says Bluetooth - but I'm using the USB cable - didn't even think to try it on my Bluetooth tablet.)

Thanks Jason 🙏🏼,

I just bought the one you suggested, plus discount for under $10.
I won’t be receiving til the 22 of October though 😢.

JasonACT · 11-10-2019, 07:07 PM

Quote:

Originally Posted by JasonACT

Oooh, that black dot wasn't there before. You will be up for a new mosfet then (I'm glad you identified what it was too - I hadn't mentioned it before because it's not a regulator, and it didn't change values for me when the screen was on or off - so whatever it's switching isn't the screen directly).

I'll do some more tracing tomorrow after work, to find where the +20v is generated from. That would seem to be the "thing" for the screen to light up (at least).

I'm not quite sure what you mean by "mv all round"?

Quoting myself.. because I did later say it is switching the screen on and possibly controls the brightness.. I'm almost sure this isn't actually true as I've done more tests this afternoon. My original comments stand I think...

It doesn't matter if the screen is on or off for me, if the 12v is applied to the large external connector for the car loom, the screen remains off and I get 21.45v on the gate pin (enabling the MOSFET). When I switch the screen on, that drops to 20.3v (still enabling the MOSFET). 12v power is going through the MOSFET and powering the large regulator, the 8v small regulator and the small unknown device - and is also sent into pin 5 of the board joining connector. The original 12v is coming in from pin 1. I'd say there's a timer that eventually switches the MOSFET off in a deep sleep mode to save power, but it doesn't happen in the first few minutes. Since you are getting dodgy voltage readings, the MOSFET still needs to be replaced before we can continue.

Pin 5 must power something else on the large board related to the screen, and I've found two components of interest (one is very close to the red and black wires).

That one is a TSSOP 16 device, I hate this one, the entire bottom is a ground which gets soldered to the circuit board as a heat sink - making it impossible to remove without damage. I just rip the legs off them and leave them there. You need to get a TSSOP 16 breakout board to use them - but I don't see any with the ground/heatsink and a heat gun might be needed to get the base soldered properly anyway. It's an LT3517 (Linear Technology) LED Driver with 1.5A switch current. It'll be doing the brightness control. They are available on ebay from China.

The less likely chip (further away and closer to the audio components on the large board) is a MSOP 10 device, I hate all of these, you guessed it, bottom is a ground/heatsink. You need a MSOP 10 breakout board..blah..blah.. Those do have a heatsink. It's got LTDXS written on it (Linear Technologies) which is their code for LT3972 - 3 Amp power regulator. Also available on ebay from China.

Would I try any of this? (Well, I have before, so that answers that.) At $700 for an exchange, I would again. $365 is sounding like an OK deal though, as I did buy a backup at around that price (which I didn't know had SAT NAV). That was also to do bench testing though, as I knew my ultimate project of getting the FGX Sync 2 unit working would require all the CAN BUS codes I could get from a MK2 unit first.

(Side note, I still don't have all the codes - I keep finding things I don't use - but at least I wouldn't miss them.)

I really hope it's just the MOSFET on the small board.

XRPete · 11-10-2019, 07:13 PM

so I bit the bullet and bought a replacement FDIM from the wreckers, ended up getting it a little cheaper too

Upon first install/startup received error message (Audio Code Error).

then...

Everything is working as it should, however the touch display seems to not work for the first minute of start up, then starts to work for some reason.

Would the touch functionality be part of the audio code error, or a different problem?

JasonACT · 11-10-2019, 07:19 PM

Screws too tight? I did mine up too tight and the plastic surround pinches on it and makes it unreliable.

JasonACT · 11-10-2019, 07:25 PM

The Menu button pretty much works right away on the bench, as soon as you can see it.

XRPete · 11-10-2019, 08:02 PM

Quote:

Originally Posted by JasonACT

Screws too tight? I did mine up too tight and the plastic surround pinches on it and makes it unreliable.

Yeah, must of been tight, just loosen then and the touch is working now..

JasonACT · 11-10-2019, 08:23 PM

I guessed correctly first time - but you might appreciate this picture too:

Damn! It should say H (igh) & M (edium).

JasonACT · 11-10-2019, 10:00 PM

An "exchanged" unit would clearly show signs of one of these components being the culprit. Just say'n. Pretty sure I private-messaged someone about the FG1 with the random loud pop/crackle issue, it would be obvious what part inside has been replaced from OEM robot built looking / soldered.

I'd be happy to pay a reasonable postage and handling fee (making it even a little bit more cheaper, hint hint - PM me).

In the words of George - let's eliminate the milkshake/ripoff.

happy1 · 11-10-2019, 10:08 PM

Quote:

Originally Posted by JasonACT

the code error goes away once Forscan has programmed it correctly. It remains properly paired until it's unplugged from the battery again.

Hi, that can't be right. If an FG ICC is coded in correctly you can remove the battery as long as you like, and when re-applying power it will be operational again. (Except for the risk of power spikes or what else causes the screen failure for MK2).

The security codes must be written to a memory chip that is not battery dependent. (It's not like a VZ Commodore where you had to enter the radio code again).

Anyway, I haven't studied the details to find out if it is the ICC that compares with a code sent from the BCM, or if it is the BCM that compares with a code sent from the ICC?

Cheers,

JasonACT · 11-10-2019, 10:11 PM

It is. The code error is a temporary thing that shows, until it confirms everything is correct. CAN BUS messages don't all happen instantly.

happy1 · 11-10-2019, 10:22 PM

Quote:

Originally Posted by JasonACT

my ultimate project of getting the FGX Sync 2 unit working would require all the CAN BUS codes I could get from a MK2 unit first.

(Side note, I still don't have all the codes - I keep finding things I don't use - but at least I wouldn't miss them.)

Hi again,
Just curious how you plan to get the Sync2 working? Is the Arduino chip able to replace messages? (Re-play certain FGX messages when another FG2 message is detected? I believe my CAN Analyser can do that, but haven't played with it yet.

There is a CAN conversion specialist in Australia that makes it possible to plug-and-play swap LSX engines into Subaru BRZ. Maybe he would be interested in developing a converter for fitting the Sync2 and Sync3 into Falcons?
https://www.agtengineering.com.au/ca...tive-solutions
Some very impressive things going on there.

By the way, If you read the reviews of Sync2, the reviews are that it is a rather poor system (comes with Windows bugs), and much better to try to retrofit Sync3.

Cheers,

JasonACT · 11-10-2019, 10:23 PM

Quote:

Originally Posted by JasonACT

You need to turn the key to ignition and the code error goes away (if Forscan has programmed it correctly). It remains properly paired until it's unplugged from the battery again.

Better?

JasonACT · 11-10-2019, 10:28 PM

I have no plan! Arduino is (to me) C++ so you can do what ever you like. Simple answer, yes and more. Yeah, the more I look at S-2 the more I'm thinking of S-3. I think it's some sort of standard though, so.. stepping stones..

JasonACT · 11-10-2019, 10:53 PM

Quote:

Originally Posted by JasonACT

It is. The code error is a temporary thing that shows, until it confirms everything is correct. CAN BUS messages don't all happen instantly.

FYI: Some don't seem to happen at all if the instrument cluster isn't turned on.

JasonACT · 11-10-2019, 11:07 PM

Quote:

Originally Posted by JasonACT

...the screen remains off and I get 21.45v on the gate pin (enabling the MOSFET).

Pretty sure this is independently repeatable, and in a court of law, could be used against Ford as exceeding specifications.

Gonads of steel may be required.

(Oh, 4 replies in a row - must be time for bed. No PMs BTW.)

XRPete · 11-10-2019, 11:27 PM

Hi JasonACT,

I did some more testing on my old board again, but not happy with my results...

Ill stop testing for now until i replace this MOSFET, and find my other multi-meter. which is better quality.

If you would like me to send my old one, PM me and we can arrange something. but ill probably keep it for a few weeks, just in case something happens with the second hand one i just bought.