Milage in ECU or gauge cluster?

[Pheonix]

it has leds for the dummy lights, but i guess that main chip is for storage as well, it leads right to the lcd.

that damn LCD is held on so loosley, i almost broke it.
it's held by a plastic case, green in this case for theme.
try to remove the case and the lcd may come with it.

on my stock cluster, i have blue leds in the back, it seems to shine through the green well.

 

[Pheonix]

thanks yahoo. hopefully the links stay.

 

[Dexter]

that big IC in picture number two is probably just the LCD driver chip....any numbers on the chips?

 

[Pheonix]

99 50 H
74HC14A

big boy
YNS inc
ICNE 10M9904B

TA 949
8008F

MC14013B


edit: some weren't correct

 

[Pheonix]

it's crazy how much room there is, you could run a whole shlew of leds throughout it.

 

[Dexter]

the MC14013B is a flip-flop logic chip.
74HC14A is a hex inverter chip.

no clue what the hell the other ones are. damn.

 

[Pheonix]

did you see the website for the big chip?
they make the whole cluster, and show the chip right on the main page and show mazda interior shots!
all japanese though

 

[Dexter]

yeah, i see that now.

and because of that, i say that it is practically impossible to reset or change the mileage yourself

 

[Pheonix]

yeah, who knows, maybe if you could get it to count really fast and roll over. they might not have even designed it to be electronically set.

you'd think mazda could set them for a replacement.

 

[TheMAN]

the links aren't working for me, can someone just attach the pics here?

 

[Pheonix]

sorry, i cannot remember what aol set my password to.

here is the yahoo photo thing.
Yahoo Photo

 

[Dexter]

yeah, who knows, maybe if you could get it to count really fast and roll over. they might not have even designed it to be electronically set.

you'd think mazda could set them for a replacement.

thats probably the only way to do it. if it even does roll over. but actually connecting up to the chip itself, is highly impossible without a datasheet or some kind of communication standard.

to try that you would have to...figure out what sends the pulse to the chip, basically, the clock line. then continually pulse it until it maxes out, and maybe it will roll over. but im not too sure of that either.

 

[TheMAN]

damn... that s*** looks complicated in design.... so why bother with chip swapping when swapping the whole damn circuit board is easier? from what I can tell, the connections from the gauges to the circuit board are larger than the chip itself, so this means it is easier to work with

 

[TheMAN]

according to my mazda documents on the BJ protege gauge cluster, the input signal feed to the speedometer is 8 pulses/one rotation of speedometer driven gear and the output signal from the speedometer is 4 pulses/one rotation of speedometer gear... it also says it runs on 12 volts... of course I would try 5 volts first so that we won't fry anything

I have no idea WTF the "input" and "output" speedometer signal is for... I'm too lazy to look in the wiring diagrams right now to see why

thats probably the only way to do it. if it even does roll over. but actually connecting up to the chip itself, is highly impossible without a datasheet or some kind of communication standard.

to try that you would have to...figure out what sends the pulse to the chip, basically, the clock line. then continually pulse it until it maxes out, and maybe it will roll over. but im not too sure of that either.

 

[Pheonix]

i'm going to make an attempt to "safely" dissasemble my DX cluster and see if the wiring is there for the tach, and maybe i could get a clean signal.

 

[Dexter]

you cant "chip swap" a surface mount soldered IC.

who knows the comm. specs of the chip and im sure YNS wouldn't just give it out.

anyways, the point isn't to trade chips, or swap circuit boards. what would be the point of that? you can just swap the entire cluster with the same results, ie incorrect mileage. the point was to figure out if the EEPROM chip that stores the information is easily accessable, and if yes, is there a way to read/write information to it. seing as how YNS seems to have a propreitary design going on here (not utilizing Microchip PIC, motorola MCU, etc), its practically impossible to "extract" the chip from the board, throw it into an MCU programmer or EEPROM writer and **** with it.
unless YNS is using one of the major MCUs and just stamping thier own name on it, but by looking at thier website, they control the entire creation of the gauge cluster, so im saying its safe to assume thier design is propreitary.

 

[Pheonix]

according to my mazda documents on the BJ protege gauge cluster, the input signal feed to the speedometer is 8 pulses/one rotation of speedometer driven gear and the output signal from the speedometer is 4 pulses/one rotation of speedometer gear... it also says it runs on 12 volts... of course I would try 5 volts first so that we won't fry anything

I have no idea WTF the "input" and "output" speedometer signal is for... I'm too lazy to look in the wiring diagrams right now to see why

it'd be crazy to see this thing run with a function generator or 555 timer.

 

[Dexter]

LOL pheonix. i was just imagining that scenario.

 

[TheMAN]

(werd) I KNOW that the odometer is on the EEPROM, *but* how it is stored on their is a mystery

you cant "chip swap" a surface mount soldered IC.

who knows the comm. specs of the chip and im sure YNS wouldn't just give it out.

anyways, the point isn't to trade chips, or swap circuit boards. what would be the point of that? you can just swap the entire cluster with the same results, ie incorrect mileage. the point was to figure out if the EEPROM chip that stores the information is easily accessable, and if yes, is there a way to read/write information to it. seing as how YNS seems to have a propreitary design going on here (not utilizing Microchip PIC, motorola MCU, etc), its practically impossible to "extract" the chip from the board, throw it into an MCU programmer or EEPROM writer and **** with it.
unless YNS is using one of the major MCUs and just stamping thier own name on it, but by looking at thier website, they control the entire creation of the gauge cluster, so im saying its safe to assume thier design is propreitary.

 

[TheMAN]

I think the first step is to dump the EEPROM and then try to reverse engineer it... maybe it uses some form of recognizable ASM code?