hwOS4: New Firmware 1.6.0 with RTE 2.09

The big one
Post Reply
heinrichsweikamp
Posts: 4258
Joined: Sunday 13. May 2007, 18:07

hwOS4: New Firmware 1.6.0 with RTE 2.09

Post by heinrichsweikamp »

ENGLISH below.

Hallo,

Die neue Firmware 1.6.0 ist nun veröffentlicht.

heinrichsweikamp.com

Anleitung zur Installation: viewtopic.php?f=7&t=19186 (Je nach aktueller Version kann es nötig sein, das Update zweimal zu starten. Der OSTC4 gibt dazu nach dem Update nötigenfalls eine Meldung aus. Grund ist, dass möglicherweise auch sehr alte Teile der Firmware wie der Schriftsatz und der Bootloader aktualisiert werden.)

Wir emfehlen ein Update über die Apps OSTConf (iOS) oder HWOS Config (Android)

Videos:
OSTConf: https://www.facebook.com/17049381164554 ... 339989145/
HWOS Config: https://www.facebook.com/17049381164554 ... 561539708/

Folgende Änderungen und Neuerungen gibt es seit der V 1.5.6:
[1.6.0] mit RTE 2.09
FIX: Automatisches Umschalten auf o2-Sensor-Ansicht:
Im Oberflächenmodus schaltet die benutzerdefinierte Ansicht automatisch auf die O2-Sensor-Ansicht um, wenn der OSTC4 auf
Sensor-Modus eingestellt ist. Dies geschah auch dann, wenn die O2-Sensor-Ansicht deaktiviert war. Dies führte zu einer automatischen Umschaltung
auf z.B. Kompass, was nicht beabsichtigt war. In der neuen Version ist die automatische Umschaltung deaktiviert, wenn
die benutzerdefinierte Ansicht deaktiviert ist
CHG: Automatische Setpoints während des Aufstiegs ignorieren:
In der bisherigen Version wurden Setpoint, die für den Abstieg konfiguriert wurden, bei der
für die Deko-Berechnung berücksichtigt, wenn ein oder mehrere zusätzliche Diluent Gase aktiviert wurden. Dieses Verfahren entspricht
nicht der üblichen Vorgehensweise, den Setpoint bis zum Ende des Tauchgangs hoch zu halten, um die Dekozeit zu verkürzen.
In der neuen Implementierung geht der OSTC 4 davon aus, dass der aktuelle Setpoint bis zum Ende des Tauchgangs verwendet wird.
FIX: Bugfix: Warnung bei niedrigem ppO2 im Bailout-Modus:
Der Code hat unterschiedliche Quellen für die Tauchgangseinstellungen (real und simuliert). Der ppo2 Berechnungscode verwendete
den Pointer auf die reale Struktur. Als Ergebnis trat im Simulationsmodus ein Fehler auf, der durch die ppo2
Berechnung, die nicht wusste, dass ein Bailout stattgefunden hat (weil die reale Struktur referenziert wurde
statt auf die des Simulators verwiesen wurde). Das Problem wurde durch die Verwendung des "stateUsed"-Pointers behoben, der
auf die aktuell verwendete Struktur verweist.
FIX: Warnung bei niedrigem ppO2-Wert im Bailout-Modus:
Der Code hat unterschiedliche Quellen für Tauchgänge (real und simuliert). Der ppo2-Berechnungscode verwendet den
Zeiger auf die reale Struktur. Als Ergebnis trat im Simulationsmodus ein Fehler auf, der durch die ppo2-Berechnung
die nicht wusste, dass ein Bailout stattgefunden hat (weil die reale Struktur anstelle der Simulatorstruktur referenziert wurde).
Simulator referenziert). Das Problem wurde durch die Verwendung des "stateUsed" Zeigers behoben, der sich auf die
Struktur verweist, die gerade verwendet wird.
FIX: Ereignis "Batterieladung abgeschlossen":
In der vorherigen Version konnte es vorkommen, dass der Ladezustand des Akkus auf 100% gesetzt wurde, obwohl der Akku
nicht vollständig geladen war. Die Ursache dafür war, dass das externe Ladegerät-IC manchmal Fertigstellung signalisierte, weil
z.B. die Verbindung zwischen Ladegerät und OSTC innerhalb eines kurzen Zeitfensters unterbrochen wird.
Dies kann passieren, während der Benutzer versucht, den OSTC in eine gute Ladeposition zu bringen. Um dies zu vermeiden,
prüft die Zustandsmaschine nun jedes Mal die Spannung auf einen gültigen Ladebereich, bevor ein Ladeende signalisiert wird.

[1.5.9] mit RTE 2.07
NEU: Ansicht "Laden" hinzugefügt:
Wenn das Ladegerät mit dem OSTC verbunden ist, wird ein neues Menü im Bereich Customview angezeigt.
Die neue Ansicht zeigt eine Ladestromanzeige, den aktuellen Ladeprozentsatz und eine Schätzung
wie lange die Beendigung des Ladezyklus dauern wird (Stundenfenster).
NEU: PSCR-Modus
NEU: Scrubber Timer verfolgt die Nutzung des CCR-Scubbers
ÄNDERUNG: Erlaubt einen Sollwert ab 0,4 ppO2: Einige Rebreather (z.B. Sentinel) starten mit einem ppO2
von 0.4 Bar => modifizierte Bereichsprüfung, um diesen Wert zuzulassen (vorher 0.5)
NEU: Neuer Menüreiter "OP" für zusätzliche CCR Optionen, z.B. Scrubber Timer und PSCR Parameter

ENGLISH:

Hello,

The new firmware 1.6.0 is now public.

heinrichsweikamp.com

Manual on how to install: viewtopic.php?f=7&t=19186

(Depending on your current version it may be required to start the update twice. The OSTC4 will show you a message after the update if this is required. Reason is, that this update also may update some older parts of the firmware like the font table and the bootloader).


To update, simply use one of our two apps OSTConf for iOS or HWOS Config for Android.
OSTConf: https://www.facebook.com/17049381164554 ... 339989145/
HWOS Config: https://www.facebook.com/17049381164554 ... 561539708/

The following changes have been made since V 1.5.6:
[1.6.0] with RTE 2.09
FIX: automatic switch to o2 sensor custom view:
In surface mode the custom view will automatically switch to the o2 sensor view in case the OSTC4 is set to
sensor mode. This was also done in case the o2 sensor view was deactivated. This lead to a automatic switch
to e.g. compass what was not intended. In the new version the automatic switching is deactivated in case
the custom view is deactivated
CHG: Ignore automatic setpoint changes during ascend:
In the previous version setpoint changes which have been configured for descending were considered
for deco calculation in case one or more additional diluents have been activated. This procedure does
not fit to the common way to keep setpoint high till end of the dive in order to get rid of the deco time.
In the new implementation the OSTC 4 assumes that the current setpoint is used till the end of the dive.
FIX: low ppO2 warning in bailout mode:
the code has different sources for divesettings (real and simulated). The ppo2 calculation code used the
pointer to the real structure. As result in simulation mode an error occured caused by the ppo2 calculation
which was not aware that a bailout happened (because the real structure was references instead of the
simulator ones). The problem has been fixed by using the "stateUsed" pointer which is refering to the
structure currenty in use.
FIX: Battery charge complete event:
In previous version it could happen that the battery charge state was set to 100% even the battery was
not completly charged. Rootcause was that the external charger IC signals sometimes completion because
e.g. the connection between charger unit and OSTC is disconnected within a short time slot. This may
happen while the user is trying to get the OSTC in a good charging position. To avoid this the state
machine now checks the voltage everytime for valid charging complete range before a charging complete
is signaled.

[1.5.9] with RTE 2.07
NEW: Added Charging view:
In case the charger is connected to the OSTC a new menu will be shown in the customer view section.
The new view shows a charging current indicator, the current charge percentage and an estimation
how long the completion of the charging cycle will take (hour window).
NEW: PSCR Mode
NEW: Scrubber Timer tracks use of CCR scrubber
CHANGE: Allow setpoint starting at 0.4 ppO2: Some rebreathers (e.g. Sentinel) start with a ppO2
of 0.4 Bar => modified range check to allow this value (was 0.5 before)
NEW: Introduced new menu tab "OP" for additional CCR options, e.g. The Scrubber timer and PSCR parameters
Last edited by heinrichsweikamp on Sunday 28. August 2022, 11:34, edited 1 time in total.
thomas.horner
Posts: 6
Joined: Saturday 29. September 2018, 11:23

Re: hwOS4: New Firmware 1.6.0 with RTE 2.09

Post by thomas.horner »

Just updated and now trying out the charging feature.
So far, it just shows:

---
Hour(s) till 100%
---

And then a chart below which shows it's charging with speed on the high-end which looks plausible.
Even after writing this text and re-checking several times, inlcuding re-activating the screen two times, it's still showing the above only - with no actual numbers only "---".

Update:
I guess after the firmware update it's necessary to "restore" the previous battery SoC first. That seems to have solved the issue for me - now it's really nice to be shown the estimated completion of the charging process 😎
Thanks for this feature!



Btw: On August 2nd I managed to accidentally "freeze" the OSTC4 again. I say "freeze" because this time it luckily didn't actually freeze or crash again, but it was "just" not updating the UI anymore and it was "only" pre-dive where I could still get it working again. To "activate" the bug, quickly deactivate+reactivate different gasses in the gas-list. Deactivating several gasses I somehow managed to be shown an empty "generic-list" and the OSTC4 didn't (seem to) react anymore. After randomly&slowly pressing buttons, after a few minutes I managed to exit the gasses screen (which it seems was still open in the background but didn't get painted to the screen) and the main screen was painted/displayed. That way the bug was gone again and I could use the OSTC4 during the dive a few minutes later, luckily. No problems during the dive 👍🏻
Last edited by thomas.horner on Friday 19. August 2022, 14:43, edited 2 times in total.
heinrichsweikamp
Posts: 4258
Joined: Sunday 13. May 2007, 18:07

Re: hwOS4: New Firmware 1.6.0 with RTE 2.09

Post by heinrichsweikamp »

thomas.horner wrote: Friday 19. August 2022, 14:27 And then a chart below which shows it's charging with speed on the high-end which looks plausible.
Even after writing this text and re-checking several times, inlcuding re-activating the screen two times, it's still showing the above only - with no actual numbers only "---".
Go to SYS2 -> Reset menu -> Maintance -> Reset Battery %

Then you could see the % rising in the charging screen.

Or wait until it's recharged to 100%. Then the new charging screen will also work.

Regards,
Matthias
thomas.horner
Posts: 6
Joined: Saturday 29. September 2018, 11:23

Re: hwOS4: New Firmware 1.6.0 with RTE 2.09

Post by thomas.horner »

Thanks Matthias, I confirm this works 😃👍🏻
thomas.horner
Posts: 6
Joined: Saturday 29. September 2018, 11:23

Re: hwOS4: New Firmware 1.6.0 with RTE 2.09

Post by thomas.horner »

thomas.horner wrote: Friday 19. August 2022, 14:27 [...]

On August 2nd I managed to accidentally "freeze" the OSTC4 again. I say "freeze" because this time it luckily didn't actually freeze or crash again, but it was "just" not updating the UI anymore and it was "only" pre-dive where I could still get it working again. To "activate" the bug, quickly deactivate+reactivate different gasses in the gas-list. Deactivating several gasses I somehow managed to be shown an empty "generic-list" and the OSTC4 didn't (seem to) react anymore. After randomly&slowly pressing buttons, after a few minutes I managed to exit the gasses screen (which it seems was still open in the background but didn't get painted to the screen) and the main screen was painted/displayed. That way the bug was gone again and I could use the OSTC4 during the dive a few minutes later, luckily. No problems during the dive 👍🏻
Version used 1.5.6, but also works with newer versions up to the current 1.6.0.
Also see attachments...

How to reproduce:

Set gasses as follows:
Air [first]
NX32 [deco]
NX50 [deco]
18/45 [deco]
Oxy [inactive]
Then, go out by pressing back. After the Logo was there, go to the gasses again.

There might be a timing component, so do the following quickly in a row:
Go to NX50
Enter on NX50
Go to Deco -> Enter
Select Deco -> Enter -> then Back
Enter on NX50
Go to Deco -> Enter
Select Deco -> Enter -> then Back
Enter on NX50
Go to Deco -> Enter
Select Deco -> Enter -> then Back

I also managed without going back - by only repeatedly pressing Enter on Deco, then Deco again, in a loop, quickly - but that might be a harder to trigger version of the bug.
Sometimes it happens that other weired things are displayed, like in the second attachment.

Note: Attachment uploading didn't work, error of the phpBB system was "Failed to move uploaded file."

The attachments can be viewed here: https://www.dropbox.com/sh/ixdfj30vwg5n ... 9I64a?dl=0
First shows the (seemingly) non-responding menu as described in the orginal post as "hang".
Second shows a first-time occurred glitch while trying to reproduce the original "hang" bug.
Just to explicitly note again: when one of those happens, it's often not possible to just return by using the "back" key, e.g. if you already accidentally changed to "edit" mode and are - invisibly, in the background - already changing the actual gas value.


Just added one more picture, which happened when I was just - slowly! - deactivating the gasses again.
So maybe the timing doesn't matter as much as I had thought. And in between, the OSTC4 had been off already, while I was typing.


If I could make a wish/proposal related to the handling:
Do you think it could make sense, if the Gas is "Deco" already and I click Deco again to deselect it, that it would be set to "Inactive"?
Post Reply