Please Help - iCharger 3010B - Micro controlled charger not react to buttons

[frasenci]

Dear All , please asking for help.

The micro starts and boot up ok.
Even managed to update (flash ) firmware.
On power up it lands on last known position of menu driven interface ( ready to charge... etc.)
None of the 4 buttons for navigating menu and selecting options react.
Have taken apart all 4 buttons, testing by jumping contacts with wire. No reaction.
PCB traces show no damage....

My question is : what would cause an apparent ok running micro to stop reacting to button pushes ( brief switch to low of pin ) ??

Can not afford cost off sending/returning to tech service and would swear I could solve it by myself..... No luck with that.

Been experimenting with micros ( Pic and Arduino ) for yeras now....

Thank you very much

Francisco
Chile

Update 1
Problem is the same before and after firmware update.
LCD screen looks updated (alive)  ok , boot up screen ok , after booting lands on same menu position and display as before the problem - after - before firmware.
Update 2
Crystal _ 24 Mhz
Posting Online Pictures here :

[pushycat]

Confirm all Vcc and Gnd are ok and no voltage drops and components heat up. Next see if any dry solder joints near the crystal. Check for ground pin missing or lose joints on the same. Re soldering the CPU pins and Xtal will give a solution at max as i guess. Next i suspected is the shorted switch one of the four. As you have mentioned that you have removed them, i didn't comment on that. Also confirm that it is now on the same condition before the flashing was done or it has changed?

[Sideshow Bob]

I had to use Google to find out what an  iCharger 3010B is. Did this error come after your software update. Can you also provide any picture of the board? That could help

[frasenci]

I had to use Google to find out what an  iCharger 3010B is. Did this error come after your software update. Can you also provide any picture of the board? That could help

It is an 1000 Watt Dc-Dc battery charger for all kinds of cells ( Lipo,Pb, etc ) Other functions as well. Used in RC Helicopter modelling .
Uploaded images link in original edited post.

Thank you
Francisco

Posted on: 09 June 2020, 17:24:37 - Automerged

Confirm all Vcc and Gnd are ok and no voltage drops and components heat up. Next see if any dry solder joints near the crystal. Check for ground pin missing or lose joints on the same. Re soldering the CPU pins and Xtal will give a solution at max as i guess. Next i suspected is the shorted switch one of the four. As you have mentioned that you have removed them, i didn't comment on that. Also confirm that it is now on the same condition before the flashing was done or it has changed?

Thank you. Will follow your tips. After and pre Flash of firmware the behaviour is the same.

I have consciously inspected the board to no avail. Will repeat with your tips in mind.
Very starnge, as it definitely not the first Microcontroller I have had at hand.

posted picture links in original post.

I am an DIY applied electronics enthusiast. Have built several real proyects. I am an Mechanical Engineer and almost ALL my electronics "know how", software and hardware, has come fom this wonderfull comunity ---> Sonsivri. It has been several years now. 15 - 20 years MG!

Thank you all

 

[Wizpic]

Looking at the photo's are the switches tactile ?

They look broke to me aren't they meant to have the tops and the push button on them ?

Wiz

[sphinx]

i surely did miss that i attached a pic for comparison.

[frasenci]

i surely did miss that i attached a pic for comparison.

Push bottoms where taken apart by me. I did jumper wire its function on testing if they where to blame. Even followed the tracks up to the micro ( micro model/brand blurred out by manufscturer ).

[pickit2]

I suspect the  problem is the transistors between the switches and main chip, It looks like it maybe a microchip pic.
I would check that the J3 program is linked to that chip. you never know they may have programmed it without setting a lock.

[witsanukai]

dose the switch wiring to microcontroller directly?
try to measure between voltage drop at each switch for compare.
If no voltage drop mean something short at that pin to ground.
May be some dust or leakage.
Try to use some alcohol to clean the board, dry it and test again.

[Sideshow Bob]

@frasenci Do you think you would be able to trace the full circutry for one switch and draw a simple schematic

[frasenci]

@frasenci Do you think you would be able to trace the full circutry for one switch and draw a simple schematic

I will !!!

Thx
 

Posted on: 10 June 2020, 08:16:36 - Automerged

I suspect the  problem is the transistors between the switches and main chip, It looks like it maybe a microchip pic.
I would check that the J3 program is linked to that chip. you never know they may have programmed it without setting a lock.

@pickit2
If I understand it right , you mean Q13,14,15,...... suspisously sitting near the buttons.... Are you thinking about inverse signal to the Micro ? Where is J3 ?
Good idea. In following @sideshow bob s idea this will result obvious after I manage to draw an schematic for the switches.
He he, I will need magnifying glass .... I will do it and post my findings.

Here I will thank all you have contributed with ideas.

Thx.
Francisco
 

[pickit2]

yes the transistors look like they are used 2 per switch.
J2 is unpopulated 6 though hole top center of the board this looks like it maybe a picchip programming port.
we use them on all our pic projects, we use a spring loaded header to field program the board.
 

[frasenci]

yes the transistors look like they are used 2 per switch.
J2 is unpopulated 6 though hole top center of the board this looks like it maybe a picchip programming port.
we use them on all our pic projects, we use a spring loaded header to field program the board.
 

The system has usb port wich is used for data logging and also for firmware flashing..
I Own an Pickit 2 Pic programmer

I am not detecting any connection between transistor pins and individual switch pins. Uploaded schematic today. Transistor detected that left pin is connected to ground , hence to common switch pins.
What should I look for ? I tend to agree with your proposition

On power, all switch pins sw1,2,3 show 4.8 volt. Except sw4.( 0 V ) Direct connect to controller pins.

On micro , have identified Vdd and Vss pins trying to identify . Does not match 18F4550 44 pin Qf chip.....I wonder.....

SW4 corresponds to the "action"- "enter" switch. On long press it causes editing off values on screen.......... it looks like the program is stuck
 right there ........"edit values" . S1 is stop. sw2 and 3 are up-down ( menu and values .) ( this when normal operation )

Thank you




 

[pushycat]

On power, all switch pins sw1,2,3 show 4.8 volt. Except sw4.( 0 V )

This point just caught my attention. If possible can you remove the switches and clean the boards LCD side with IPA and physically see for shorted tracks or the sw4 itself may be. Trace the common track to all the 4 switches. Next trace individual tracks from 4 switches to microcontroller where all should give you same voltage as i guess. As you say sw4 shows 0, i suspect something fishy there. Also what lies below the LCD? If LCD is removable, please  see beneath it too. I also strongly suggest that you can see if anything changes with the replacing of the Crystal.

[frasenci]

This point just caught my attention. If possible can you remove the switches and clean the boards LCD side with IPA and physically see for shorted tracks or the sw4 itself may be. Trace the common track to all the 4 switches. Next trace individual tracks from 4 switches to microcontroller where all should give you same voltage as i guess. As you say sw4 shows 0, i suspect something fishy there. Also what lies below the LCD? If LCD is removable, please  see beneath it too. I also strongly suggest that you can see if anything changes with the replacing of the Crystal.

LCD soldered , could remove but nothing fancy there ( except for a little dust...) Last to do....

I wonder about the crystal ( 8 Mhz ) ..... if faulty , the LCD screen would not work , dont you agree ? Just wondering. Could replace ok , some crystals lying around...
Cleaning and backtracking right now , he he
Thank you

[pushycat]

I wonder about the crystal ( 8 Mhz ) ..... if faulty , the LCD screen would not work , dont you agree ? Just wondering. Could replace ok

As I dont have the board in hand and we have nothing as reference schematic, all we can do is just a guess work with our grayed hair. So as my experience goes, i have seen corroded Crystal pins and faulty crystals over a period of time which may not work up to the mark. Once powered, the oscillation may start and dampen later which will cause the system to freeze. So just trying it with a new crystal will just clear a doubt on that part.

[frasenci]

with our grayed hair.

What do you mean by that ?? I am just over 62..... he  he he   

Greetings
Francisco

[Sideshow Bob]

As I dont have the board in hand and we have nothing as reference schematic, all we can do is just a guess work with our grayed hair. So as my experience goes, i have seen corroded Crystal pins and faulty crystals over a period of time which may not work up to the mark. Once powered, the oscillation may start and dampen later which will cause the system to freeze. So just trying it with a new crystal will just clear a doubt on that part.

But if he/she was able to flash the charger. Should not that be a strong indication that the crystal is OK

[comlekciler]

I think;

grayed hair = years = experiences 

[pickit2]

if the display is working but the buttons don't then there are a few things that would give this condition.
1. the input ports are damaged or the switches, (I think the transistors are connected, as it's unlikely the switches would go to the main ic, direct, bad practice if no protection of the inputs
2. the microprocessor program has been corrupted, and unable to load the main program in full.

maybe due to a spike from the power supply as there is a high current. also the transistors between the switches if not part of the input, would be very bad practice, there is a lot of room on the board so why place them there.

[frasenci]

if the display is working but the buttons don't then there are a few things that would give this condition.
1. the input ports are damaged or the switches, (I think the transistors are connected, as it's unlikely the switches would go to the main ic, direct, bad practice if no protection of the inputs
2. the microprocessor program has been corrupted, and unable to load the main program in full.

maybe due to a spike from the power supply as there is a high current. also the transistors between the switches if not part of the input, would be very bad practice, there is a lot of room on the board so why place them there.

1.- Made macro picture from transistors area. Will carefully track therm . I already checked "left" pin = Gnd and can see "right" pin to micro port pins (other than those of switches ). Third pin ( base ? ) not clear.

2.- Could be , last use for charger was charging batterys from Solar panel source. This "old" charger has some notable MMPT tracking abbilities matching perfectly the imput Voltage to the MPOV of the panels.
Was left connected overnight and repowered at sun rise (no switch ) waiting just to hit the start button (sw1 ) Sounds weird but its original destination was to be hooked up to a DC source such as car  battery ( always a spark at connection ---- > boot up.
Several months in this routine. One morning the switches are dead. Charger always indoors.....)

Q: Should I flash firmware again ?   ( i.e. corrupted code ) Although pre/post flash misdbehaviour was the same.


Thasnk you Sir
Greetings

Board details "pickit2 region" 
https://pasteboard.co/JcuCqi1.jpg
https://pasteboard.co/JcuCP1l.jpg

[pickit2]

J6 is a Npn so left pin2 = e  right pin1 = base pin3 = col
I can remember a panel I worked on when the keypad was only in use when setting alarm at known time or condition.
the keys were enabled via a transistor they were disabled after alarm was set until the unset time. maybe that is in your code to. or transistors are shorted to ground by an internal short, due to flash over. it looks like the transistors are in pairs. did the unit have lock out while charging?

as to flashing did you flash with the right version for the device. you may have flashed a newer version than your unit build.


Posted on: June 11, 2020, 02:08:46 02:08 - Automerged
looking at the main chip can you post a side view looking at the side of the chip and the pads.
your chip may have over heated and moved on the pcb

[pushycat]

What do you mean by that ?? I am just over 62..... he  he he  

Greetings
Francisco

Then you have white. Mine is just grayed..
Did you check the crystal and its pins? Re-soldering the main chip with fresh solder may come out with a good result.

[b555b]

Is that device also remote controllable through its USB interface, or is it only used for firmware update? If it is remote controllable, it could be a good way to check if the problem arise because of a possible problem with the buttons, or because of a problem with the firmware crashing after having displayed the main menu.

[frasenci]

J6 is a Npn so left pin2 = e  right pin1 = base pin3 = col
I can remember a panel I worked on when the keypad was only in use when setting alarm at known time or condition.
the keys were enabled via a transistor they were disabled after alarm was set until the unset time. maybe that is in your code to. or transistors are shorted to ground by an internal short, due to flash over. it looks like the transistors are in pairs. did the unit have lock out while charging?

as to flashing did you flash with the right version for the device. you may have flashed a newer version than your unit build.


Posted on: June 11, 2020, 02:08:46 02:08 - Automerged
looking at the main chip can you post a side view looking at the side of the chip and the pads.
your chip may have over heated and moved on the pcb

For your consideration...



Flash version as recommended by manufacturer ( Junsi Charger ) - Behaviour same before-after

Keypad asumed diferent actiosn depending on state of charger - All 4 switches always active

Thank You


Posted on: 11 June 2020, 14:06:38 - Automerged

Is that device also remote controllable through its USB interface, or is it only used for firmware update? If it is remote controllable, it could be a good way to check if the problem arise because of a possible problem with the buttons, or because of a problem with the firmware crashing after having displayed the main menu.

USB only output ( 1 hz data train with all variables...... Includes a Temperature sensor ( analog presumable LM35 type ))
This I researched because I wanted to remote control it ( Start/Stop) via ESP8266 Wifi Serial out or something.. No luck

I was in the process of replicating the switches precisely bringing Sw1 line to ground per ESP8266 Wif module.... just a proyect , I did not have the chance, although I checked that this was the switches logic ( brief bringing to ground )

[sphinx]

To me that looks a bit of not so good soldering, but it is not what makes me wondering. It is the white/grey residue on the board could, that be from water/moisture or is it perhaps
from the cleaning of the board when it was made. From my experience moisture over time can make things go bad if the board is not protected from corrosion.
just my input.

[Kombinator]

Try forcibly raising the pin potential of button 4 with a limiting resistor to 20mA (+ -> 250 ohm-> pin controller).
Good luck.