@Boardsurfer : No, it was not intentionally, but again a timebug.... It was already in SW 5.68.
@Remery and Stretchy : Probably the same timebug, I am working on a update.

SW 5.70 is the next attack against the time bugs !! I followed the advice from Peter, time is now set when logging start. You can follow this on the screen :
Wifi Off S2
Waiting for Sats
and then
Wifi Off S5
Waiting for speed
and then when logging starts, Stat or Speed screens.

When now the min speed is exceeded, time is set and logging is started. Added another check for not plausible dates.
SW5.70
e-paper keyword "now" may not be used for getLocalTime() !!
Same issue in SD_card.cpp -> open files !!
GPS_SAT_INFO buffer 16 -> 10
Set GPS_time @ start logging
check for year>2023
While waiting for Sats / Min speed : Wifi Off with waiting Sats/Speed on screen
Total time in .Txt from start logging
Greetings Jan.
github.com/RP6conrad/ESP-GPS-Logger

Thanks for your great work Jan

Just uploaded sw5.7. One unit went fine, the other now has a corrupted config file, it will no longer connect to wifi.
As it is sealed and there is no access to the USB port, Is the unit now US?
Can I recover from this somehow?
I've just tried Ap mode, the device says FTP2, but I can't connect with browser or FTP client.
2nd window on boot Config Fail
3rd window on boot ssid NOK. Which is nothing like what I put in.
sleep display shows "MY ID"
display in KM/hr still works. So I can still use it on the boom for on the water feed back, but I have to do a mental calculation to get back into knots.
I don't think I'll do another update on the other good unit, as I need ma back up incase my head unit fails.

OK all fixed, I'd forgotten I needed to change from router to unit in the computer wifi

That's a good solution Jan. I worked in a phone app that recorded user location. Sometimes the fix was poor and the location was out by 60km! So we left the GPS running in the background while the user filled in the form, this gave time for a decent fix. We were probably recording date and time off the phone so never encountered corrupted times.

I just walked around the block with 5.70 and date and time were good.

You might want to look at the colon alignments on the standby screen :)

out of interest has anything changed with the reed switch?
turning on is still fine, but turning off seems to take a few attempts?

Mine did the same, almost like it was frozen for a couple of minutes and then was fine. "We're gonna need a bigger magnet!"

As Jan posted the start up is now different.

You get a pause screen till you start your first run then all is good.

Dates seems to have been sorted for my first session with 5.70

Select to expand quote

decrepit said..
OK all fixed, I'd forgotten I needed to change from router to unit in the computer wifi

You know what they call this M ?. I an IBM error (Idiot behind machine error)

Just noticed that the .gpy time bug is back again . My fault, with going back to the old sw for fixing the slow update of the e-paper. No big deal, as this is a known fault. Will be fixed with the next update.

Select to expand quote

Rolz said..
out of interest has anything changed with the reed switch?
turning on is still fine, but turning off seems to take a few attempts?

Nothing should have changed there, but you never know.... Time is still 1500 ms. See no difference on my units., I will check it.

I used both of my units on AP mode today. One of my units wont shut down properly and is still showing the Wifi on AP screen, but all lights are off. Software is V5.67. Any ideas?

If nothing is connected to the AP, after a while Wifi Off should appear, and you can switch it off.

Select to expand quote

rp6conrad said..
If nothing is connected to the AP, after a while Wifi Off should appear, and you can switch it off.

It has been a few hours already

What is the bat voltage ? Did you you try to charge it ?

Select to expand quote

rp6conrad said..
What is the bat voltage ? Did you you try to charge it ?

at first I was reluctant to charge it in case I was going to have to let the battery run down in order for it to reboot, but when it was still stuck on the same screen for many hours I thought I might as well try charging it. Batt voltage was around 4, now 4.15

Will it turn ON? That's the most important thing.

Select to expand quote

decrepit said..
Will it turn ON? That's the most important thing.

no. it is frozen

A stronger magnet?

Select to expand quote

remery said..
A stronger magnet?

I tried a super magnet

Select to expand quote

Windxtasy said..
at first I was reluctant to charge it in case I was going to have to let the battery run down in order for it to reboot, but when it was still stuck on the same screen for many hours I thought I might as well try charging it. Batt voltage was around 4, now 4.15T

The good news is that the program is still running, as you have a update of the Batt voltage. The bad news is that you have no lights on the GPS and no reaction on the reed switch. I had a unit with water ingress, the GPS died and the lights were out. I had another unit with a bad soldering on the reed switch, but with some shocks, it worked again for a while.
Not much options left : wait for the battery to die, and try then to wake it up again.
Greetings, Jan.

Select to expand quote

rp6conrad said..

Windxtasy said..
at first I was reluctant to charge it in case I was going to have to let the battery run down in order for it to reboot, but when it was still stuck on the same screen for many hours I thought I might as well try charging it. Batt voltage was around 4, now 4.15T

The good news is that the program is still running, as you have a update of the Batt voltage. The bad news is that you have no lights on the GPS and no reaction on the reed switch. I had a unit with water ingress, the GPS died and the lights were out. I had another unit with a bad soldering on the reed switch, but with some shocks, it worked again for a while.
Not much options left : wait for the battery to die, and try then to wake it up again.
Greetings, Jan.

Voltage hasn't changed since ths morning. That is going to take a long time.

Rp6conrad (Jan)... request:

In the webinterface -> files: if i want to delete 1 file, no problem, its a bit slow but its 1 time. Now i want to delete more files, than it takes to much time. Can you make a methode to delete multiple files at once (perhaps wilt checkboxes)?

And i recognise the turning off issues......

Select to expand quote

Windxtasy said..

rp6conrad said..

Windxtasy said..
at first I was reluctant to charge it in case I was going to have to let the battery run down in order for it to reboot, but when it was still stuck on the same screen for many hours I thought I might as well try charging it. Batt voltage was around 4, now 4.15T

The good news is that the program is still running, as you have a update of the Batt voltage. The bad news is that you have no lights on the GPS and no reaction on the reed switch. I had a unit with water ingress, the GPS died and the lights were out. I had another unit with a bad soldering on the reed switch, but with some shocks, it worked again for a while.
Not much options left : wait for the battery to die, and try then to wake it up again.
Greetings, Jan.

Voltage hasn't changed since ths morning. That is going to take a long time.

2 days now. Still frozen. Batt voltage/ display hasn't changed

It's probably not updating the battery voltage. Bummer you won't know when it's flat. I've no idea how long you'll have to leave it.
I have a vague memory of Somebody saying their unit came back to life after a long delay, may have been Alby.

If the controller has gone off the rails and sis till running it shouldn't take too long, but if it's just shut itself right down could take ages.

I am working on a update for file handling : The files will be sorted and the newest files will come on top. For deleting files, I suggest that I make a setting for "delete alle files older then xxx days" (and younger then 2021....).
For now, the most convenient way for deleting files is with FTP (I use WinSCP).
@Windextasy : I am afraid that there is a hardware issue. Normally, the unit is in "standby", with the standby screen. The it wakes every 3000s to measure the batt, and refreshes the standby screen. Or it is in Run, with al the other screens. Or, there is no voltage anymore, and the ESP32 is then completely switched off. It can be that the last screen will stay on.
For the reed switch issue : I cannot reproduce it with my units, they switch off normally. You could go back to a older SW5.66 version, to see if it make any difference.

Greetings, Jan.

Just wanted to share a small win... got an esp32 with a non-responding SD card.
Unfortunately it's now Chinese New Year so any response from Lilygo was going to be a few week off...
Decided to remove the onboard SD card and then connected an external one and it all worked.
So impressed with the esp32 boards.

For dealing with these issues (DOA, death on arrival), I changed the workflow in the build manual : As the firmware on the virgin T5 board checks the SD-card, and report the free space, problems here can be seen before soldering / flashing the unit. It is always better to battle on 1 issue at the time !
DIY-gps build instruction :
docs.google.com/document/d/1XJD6BV1Xay4nXo0cDuo_IOtxC_f1LyG7DE7X6KGlAB0/edit?usp=sharing
Greetings, Jan.

Select to expand quote

rp6conrad said..
For dealing with these issues (DOA, death on arrival), I changed the workflow in the build manual : As the firmware on the virgin T5 board checks the SD-card, and report the free space, problems here can be seen before soldering / flashing the unit. It is always better to battle on 1 issue at the time !
DIY-gps build instruction :
docs.google.com/document/d/1XJD6BV1Xay4nXo0cDuo_IOtxC_f1LyG7DE7X6KGlAB0/edit?usp=sharing
Greetings, Jan.

yeah next time I'll be testing board first before assuming all is good and get into the build.
rookie error