Help ID a MCU from failed china made speedo

[darkspr1te]

Hi All,
So my story is, I bought a bike from a friend last month and he said the dealer had replaced the speedo with a new one as old one failed to display anything but fuel and indicators , this was around june and they took it off a new display model and fitted it to the bike in front of him and it worked for a while for me after buying the bike but then failed in same way.


I have already replaced it with a American market hawk 250 lcd display but was wondering about maybe fixing the old one too.


Here is where i need some help from all the mcu experts out there, it does not seem like a stm product which is the only one am familiar about and a casual check on atmel chips does not match the pattern.
attached is a picture that show it's xtal pins, spi/i2c , programming / major io and power /adc side but cant seem to match pin 1 , there no dot and no corner shape difference i can see with my eyes


it's driving a old school lcd style display like a watch for speed/rpm/gear/ trip from the number of pins and not a modern lcd style spi/data i think
oddly the build date is 20230507  for this replacement taken off a new bike :-P
shown in pic two
so not so good QC going on at this factory


I will power it up on the bench and see what the output do on a scope, it does not seem like the mcu has failed as many functions still work like speed , trip , gear but rev counter and efi light do not come on.


darkspr1te

#edit# corrected build date  *

[fzabkar]

"oddly the build date is 20030507"

I'm seeing 20230507. :-?

I would think that the speed sensor would be an open collector Hall Effect device, in which case the input to the PCB would have a pullup resistor to the supply voltage, possibly +5V. I would also think that there may be an ESD protection component on this input.

Could SP+ be the supply, and SP the output of the sensor?

[Kim Christensen]

Looks like you might have some solder bridging around that IC. Don't recognize the MCU but:

[Gyro]

The problem isn't just the id of the mcu, it's that it will have been programmed too. A blank new one isn't going to do you any good without the code.

Agreed, the soldering is lousy. Combine a bad factory soldering job with no-clean flux, vibration, flex and maybe a little moisture and it could be the source of your problem. It would certainly be worth cleaning the flux off and touching up the solder joints.

[darkspr1te]

Yeah, the soldering is bad, dry flux all over, solder blobs , bridging.
I have already cleaned it with IPA now and using soft brush knocked off all of the solder balls and that seems to have fixed most of the display but some items still dont display.
Am not worried about the existing code, once i find out the mcu it will be my next job to figure out a way of dumping it, I already have ways into pic 8's, stm8/stm32s , qualcomm chips to dump the flash and writing code for a simple dash display is no biggy either, would be fun actually to hide easter eggs in there.
The device has only one button for changing display of trip/odometer and my boot up hold button down tests have given me what maybe a service menu but with some of the display not working it's hard to tell. The replacement does not trigger in this way on the bike, it seems there has to be a additional B+ signal that my bike loom does not have but my bench loom does.


Currently waiting some more solder braid so i can clean up the excess solder and then run the fat tip over all the legs with some quality liquid flux then a quick IPA bath
Oh thanks for pointing out my date typo, have corrected the original now.


Speed sensor is hall effect, the factory sensor is all in one unit encased on the wheel strut with a magnet on the rim , its a 5v type so ground, 5v+ and signal, I actually got a extra one with the Hawk 250 replacement so i will be able to test the speedo with a card wheel+drill on my desk or grab a 5v arduino for a false signal generator.


One of my concerns was ensuring the odometer is correct as i take my service schedules serious , my father was a mechanic and drilled it into me to service regular. as so far it seems the prior owner and service center just went "meh a zero value is fine, done less riding right ? " but thankfully i had photo's of odometer just before each one failed so i have a within 500km figure.


I think tonight i will sped time on github looking for mcu schematics that use these pins as well as test what type of IO that programming port is, 2 wire says to me SWD or plain UART or a rare case of jtag over swd. None of my usual chineese tech sites that disassemble stuff have featured these boards (quite common over there on their bikes, just watch any bike video and you will see loads)




darkspr1te