Display only show up after softreset

e8abac78cdf65c56

-J15_S: jumperd to 3.3V
Voltages: from left to right, when looking at the loopa the midi-sockets are on top (J15 label is on right), Ground as base:

3,3MOhm to ground (when turned off), 0V to ground when turned on - Connection to 595 OK
3,28V
0,34V @ 1,1Khz (says my flucke 115)
0,35V @ 1,1 Khz
3,08V
3,2V
Ground

-Connection: 595 > Waveshare = OK

more soon

e8abac78cdf65c56

-R2: 1K
-C4: 10uF 50V polarized is ok, when removed it is the same behavior (stay black until, reset)
-D4: Orientation is ok. No short, measure 0,5V in booth direction with the diode tester (while part is soldered in, and there is no power
-R33A-C: 1K, D: 550Ohm > OK

595 to J15C and GNd+3.3V all connections > OK

R32 and R19 are 100 Ohm — connection to MCU <>J15 Pin 2 > OK

about waveshare > OLED wireeing i need some assistance (shematic)

17703227d9536da1

As written, OLED wiring follows the STM32F4 schematic:

net N$9 connects to IC2, pin 15.

Did you check the J15_S jumper?

17703227d9536da1

Newhaven display (from datasheet).

e8abac78cdf65c56

EDIT: i made a error in the shematic
please delete my last 2 posts

yes DB3-7 are on ground and pin 9 is measured 5,5MOhm to ground

The connection from the pcb - J15 (solderpoints) and OLED-Screen - Solderpoints is ok

OLED-Measurment:

PIN8 (OLED) > PC11(MCU): 0 OHM > OK
PIN7 (OLED) > PC9 (MCU): 0 OHM > OK
PIN4 (OLED) > PA8 (MCU): 0 OHM > OK

J15C core pcb > OLED-Solder Points >>> 0 OHM > OK

17703227d9536da1

Not sure what to suggest then, connections seem okay. So maybe the OLED receives some junk data on startup, which puts it into an undefined state?

This is the purpose of the POR (power-on reset) circuit: it delays the power seen by the reset signal. Maybe you need a longer delay? Try a larger capacitor?

Other ideas are: noisy PSU or flux residue around the 595 chip, or between headers/MCU pinheaders.

e8abac78cdf65c56

i replaced now R2 and D4 with new parts (and cleaned everything under and beside it)

without C4 - not working
0.68uf ₀.5s to 3.3V not working
4.7uf ₀.5s to 3.3V not working
10uf ₀.5s to 3.3V not working
100uF _1s to 3.3V not working
220uF _2s to 3.3V not working
2200uf ₃.5s to 3.3V = not working

when hit the reset button, i see the progress bar of the looper (where you see notes usw) for a short time, then the loopa-logo apears. is this the loadet data from the state bevore i reset - or is that normal?

e8abac78cdf65c56

following thing is working:
remove jumper j15_s
connect usb power
wait at least 2 seconds (this is the time LED1 NORMALY flashes 2 times, but it will not flash in this case)
set jumper j15_s (3.3V of course)
Now Loopa boots as normal

e8abac78cdf65c56

A reset via Mios-Terminal, work as well as the Reset-Switch on the backside off the device

what else done: (without any change of the problem)

used different power supplys
soldered a 2200uF 16V cap between the PA9 VBUS and ground
100nf Kerko (SMD) between Ground and 3.3V of the Display
FLASHED the MCU with the STM utility with the bootloader V1.018
tryd fastboot 1, or 0….
used then fastboot 0 (because thats the virgin bootloader setting)
after uploading the looper 2.08-file, on the first start, the screen apeared (its something like softreset….)

so next thing i will try is to remove IC2 (removing is more complicated then soldering…) - but that tomorrow…

17703227d9536da1

Seems like the display init is not working somehow. Could be related to IC2 I guess. It is always worth soldering and checking for shorts on any pins that carry display data.

You could also try a direct connection of 3v3 to the display from the waveshare board. You’ll also notice a spare pin on the second display connector (pin 2, to the right of the square one, please verify the connection to the 3v3 on the display first).

e8abac78cdf65c56

Hawkeye

tryed your idea (again - i had the same…), after the loopas-hex-file-upload i see the screen (because of soft reset), but after plug in and out the usb again, it stays black… The J15_S trick, to set the jumper 2secs after powering up, now works with 1sec.

latigid on

direct wire to waveshare make no difference (incl. mass and 3.3V)
the resistance between neighbar-pins on IC2 is generelly 1,5 to 8MOHM except Pin13&14 (RW&RS: 1,5KOHM) that is i think normal because they are connected to the resistor network.
i scraped between the pads (which where allready clean), cleaned, give bathes - all with no change
now i sadly remove the IC2… lets see

0fc5608d995b44ff

@Phatline something to try:
a) install MIOS32 bootloader app for STM32F4:
http://ucapps.de/mios32/mios32_bootloader_v1_018.zip

b) boot bootloader app

c) Within MIOS Studio click on the command/console line and type:

set fastboot 1
store

d) now the bootloader does not wait on MIDI IN/UART commands in the first two seconds after startup anymore.

e) flash LoopA v2.08 firmware and try again

If this works, you may have stray inputs from any UART (MIDI IN 1,2 or 4), which causes the bootloader to wait for an app upload via the classic MIDI ports (which then never happens) - that would explain what is happening.

Best regards,
Peter

e8abac78cdf65c56

it was IC2, or the stuff under it…

First i removed R33B to get place to work with the Soldering - Iron
I removed IC2 Leg by Led with a Tweezer (de: Pinzette), > heating the Pad, then force the Leg up
After IC2 was gone I desoldered all Pads, and cleaned the area under IC2, and of course around It (Isopropanol)
Soldered a new 74HCT595D (Philips)
Soldered R33B
connect the USB-Power - TaDa.

17703227d9536da1

Well done, glad that it is working!

0fc5608d995b44ff

Good job! It could have been the IC - believe it or not, two weeks ago i encountered my first faulty 74HC165 in life 🙂.

Many greets,
Peter

17703227d9536da1

BTW @Phatline

If you don’t already have a hot-air reflow station, I would strongly recommend one. You won’t necessarily be able to reuse the chips after desoldering and have to be careful of overheating the surrounding parts, but it normally takes just a few seconds to completely flow all pads, then slide the chip off (I prefer to do it along the long axis of the pads to reduce mess and track lifting).

Sometimes also a bit of hot air around the leads will bake out the flux or separate small bridges.

17a82e0128a6f9fe

well, really interesting thread. suddenly I have part of my reponce here.
the J15A and J15B have the same distribution pins.

I wanted to use another OLED LCD display on the J15B connector (to the left of the display) of type SSD1322 but I do not know the type of driver of the official display (connector at the top) it is also an SSD1322?
if not, is it possible to change the parameters to change the display type?

17703227d9536da1

Both of the headers for displays on the LoopA Plate PCB are suited for SSD1322. LoopA will also work with the “White 3.2 inch Arduino,Raspberry Pi OLED Display Module 256×64 SPI” on BuyDisplay.

0fc5608d995b44ff

@phd0 please note, that LoopA’s software will only support a single SSD1322 - if you want to use your LoopA to run a different app (uploaded via MIOS Studio), you can use any of the supported MIDIbox display types - have a look here: http://ucapps.de/mbhp_lcd.html
Best regards and enjoy!
Peter

17a82e0128a6f9fe

thanks everyone.
i look some Kicad screenshot a the top, but i don’t find theses files on the ucapps site.
i searche it for the SEQ 4+ too.