Maple mini from BAITE – the real PILL

i really appreciate it when other people report my insanity as theirs.

i am no longer alone.

stephen

The blank screen thing sounds familiar, I messed around for a couple of hours trying to figure out what was wrong with one of my boards, which had been running just fine, then stopped. I double checked all the code, tried different pins, all the usual trouble shooting tricks.

The cause was poor soldering on the display board. The ribbon cable connecting the display to the board had partially separated from the PCB. A little microsurgery later and it was all working again.

I too was beginning to question my sanity… however it turned out that my tenuous grip on reality was not the cause of the issue after all

andy,
why would you change the software and change pins/interfaces within what was a working setup?
working state to non-working is a hard failure, unless M$ anything is in use.

@bianchifan
that sounds more like software

awhile back i had fun with ili9341 displays, eventually i ‘found’ a working combination of software, wiring, maple mini and display.
i labeled the display and the maple, cycled through all the ili9341 displays i have. all worked.
swapping back to the original display, i went through my maples, surprisingly only one failed.
then i got a working setup together on a pill, then did a pass with all the pills.

having a bread board with a 40 pin ZIF socket is rather useful
i’ve done this exercise on my nokia displays, i2c and spi small displays as well.

now i have a maple mini and a couple of pills with all their pins folded inward and flattened.

since then i’ve had some real fun with things not working, seemly rather insane, they seem to have disappeared with copious operation of the reset button, also disconnect/reconnect as well.
for pills, i always use stlink and for maples bootloader 2, never serial; for other boards also stlink to the appropriate pins.

but with a working setup suddenly non-functional :-
1. check connections – anything on a cable/ribbon – both ends electrically and visually – oc / sc / adjacent.
2. attached hardware(s) – individually on another board if poss – using the same software if possible, map to an alternative i2c or spi if not
3. connect another display if you have one
4. if possible, replace the board(cpu).
change software – no

whilst the hardware/software interface has migrated towards the latter, debug wise on a working setup
it’s much more likely hardware that has gone astray.

stephen

zmemw16 wrote:andy,
why would you change the software and change pins/interfaces within what was a working setup?
working state to non-working is a hard failure, unless M$ anything is in use.
……
stephen

@andy
apols, that was a bit of a broadside.

luckily i still have hair to pull out. sanity – not much and thinning.

i still have this feeling that something is not right. if i’m getting consistent behavior then i can set about finding it. inconsistent behavior is more difficult. i’d never be employed at Bletchley Park.

on stm32, spi always seems to be a bit of a struggle, i2c also; at least to me, worse if doubled up( spi 1+2 or i2c 1+2.)
using spi or i2c and a nano or uno, they work.

why does a digital ‘system’ not behave in a deterministic manner, timing, race condition?
so is a maple mini programmed with a sketch deterministic?

stephen
now looking for his crash helmet, yet again.

When things start behaving oddly, not resetting, not booting and so forth, I tend to suspect the power supply first. Having said that, I recently built a simple opto-isolated serial board for my multimeter, from one of those cheap $2 Chinese usb serial boards and a photo-transistor recovered from an opto-isolator. It worked fine until I tried to assemble the whole thing in its completed form.

At that point…. nothing, not a flicker of life.. took it all apart, and started to test it, and it started working again. Much head scratching ensued..

The issue turned out to be my assumption that a single layer of insulating tape was sufficient to keep the PCB from shorting out on the photo-transistor.
Image
A little pressure on the board when it was in its case caused a sharp leg of the through hole 12 MHz crystal to pierce the insulating tape and contact one of the legs of the transistor.
Image

Hot glue and 2 layers of tape fixed the issue. What little remains of my sanity was preserved

andy,
i use those bread board psu units fed with 9v or 12v, i apply a 1 hole offset/twist to them, hopefully it might improve the connections.

is my only seeing a no entry bar something i’ve missed in my 32 to 64 bit OS shift or at your end?

i cheated for pulse detection with ldr modules at about 60-65p, so i got a few, up to what free postage allowed me to! oops 33p on aliexpress, my orders says i got a pair and a quartet of them.

i suspect it”s an adjustment pot and a comparator with digital out, seems to work on 3v3 quite happily.

aside from that, max freq for a single analog channel would be – guessimate would do, arduino style & as direct as poss. any idea?
i’m wanting to capture approx 50mS pulses and wanting to see their edges. electricity meter 1000p/kw type.

further than that, feline eyes, definitely fully reactive, unequal pupils – natural for her or similar causes as in humans?

stephen

ahull wrote:The cause was poor soldering on the display board. The ribbon cable connecting the display to the board had partially separated from the PCB

zmemw16 wrote:andy,
i use those bread board psu units fed with 9v or 12v, i apply a 1 hole offset/twist to them, hopefully it might improve the connections.

excellent link for the feline
i’ve been putting her psychotic moods down to her 1st 6wks, the last of those weeks is apparently the ‘how to behave with humans, e.g. handling, tickling, stroking, playing’; reflects the personality of her then owner.
her behavior is consistent over last 3-4 yrs – so i don’t think it’s a real issue.
she’ll snap teeth or claws at you for seemingly nowt, it’s certainly improved my reflexes.
to be fair it is a snap or a bat; first time at least.
btw i think^H^H^H^H^H^H^H the ‘dreamie et al’ cat treats are addictive.

that dead picture, can you post the link as text or what its format?

stephen

zmemw16 wrote:excellent link for the feline