Repository navigation
Beta v0.2.0
opendash 0.2.0 — a session that survives the drive, and a list that tells the truth about the car
Context: three days, 15 to 17 September 2026, on top of
v0.1.0-beta. Ten commits
inopendash, two inopendash-plugins, four inscanmatik-bench. One session at the
vehicle — an Opel Astra J 1.7 CDTI (A17DTJ, Denso ECU) — driven from a Windows bench
over the adapter's Wi-Fi while the owner worked the pedals.Theme: 0.1.0 established that the app could talk to the car. This release is about
everything that was still wrong once it could: a recording that came home empty, a
viewer that could not open its own files, a pedal reading 798 %, injectors reported as
broken that were not, and a forty-seven-parameter selection in which thirty-seven
parameters never received a single sample.Deliverable:
android-arm64-v8a-release.apk, versionCode 2, versionName 0.2.0,
minSdk 26, three languages (en/es/de). 50 Kotlin files, 11 Rust files, 242 strings.
1. Change catalogue
| ID | Area | Item | Type |
|---|---|---|---|
| L1 | Link | Channel narrowed to four PASS filter ranges instead of the whole bus | fix |
| S5 | Streaming | All seven packets used, not five: fourteen identifiers, not ten | fix |
| S6 | Streaming | What does not fit in the packets is polled rather than dropped | fix |
| S7 | Streaming | Frames per second reported alongside samples per second | feature |
| S8 | Streaming | Extras polled every 300 ms instead of once per loop | fix |
| D1 | Data | Row width taken from the car's answer, not from the catalogue | fix |
| D2 | Data | Signed rows read as signed where the declared minimum is below zero | fix |
| D3 | Data | Unitless single-bit rows read as On/Off rather than 1.00 | fix |
| R2 | Recording | Foreground service, Wi-Fi lock and wake lock, so a locked screen keeps recording | fix |
| R3 | Recording | Recording on by default, and next to Start rather than folded away | fix |
| R4 | Recording | Reader takes its columns from the header, so it can open what the writer writes | fix |
| R5 | Recording | Flush every 250 ms rather than every row | fix |
| R6 | Recording | No file created until the first row, so an empty run leaves nothing | fix |
| U1 | UI | Live screen is one scrolling surface; the setup panel is the list's first item | fix |
| U2 | UI | Selected (n) filter, and named groups per module | feature |
| U3 | UI | The fast switch says why it is refusing | fix |
| U4 | UI | System back closes the screen you are on, not the app | fix |
| C3 | Catalogue | A catalogue can publish profiles.txt: named selections, ready-made |
feature |
| C4 | Catalogue | A catalogue carries a content revision, so refresh can report an update | feature |
| C5 | Catalogue | Profiles are listed and importable from the catalogues screen, without the car | fix |
| A10 | Actuation | Rows naming an output are listed with a padlock; the Settings lock now gates something | feature |
| A11 | Actuation | $AE DeviceControl permitted only while unlocked; AE 00 release implemented |
feature |
| B1 | Bench | 63-perfil-del-coche.mjs builds a car's profile from a probe |
feature |
| B2 | Bench | 82-grafo-del-vatbin.mjs reads GDS2's option tree with its own loader's format |
feature |
| B3 | Bench | 83-actuadores.mjs inventories what the ECM drives |
feature |
| B4 | Bench | 60-app-en-windows.mjs --hacer maniobras: a time series, driven from the bench |
feature |
2. Measured at the vehicle, 17 September
Engine running, idle to 3 000 rpm, the bench driving the adapter over Wi-Fi from a
Windows machine while the owner worked the throttle.
| Measurement | Value |
|---|---|
| Unfiltered bus, this car | ~1 340 CAN frames/s |
| Of those, frames with the exact shape of a complete ISO-TP single frame | ~173/s |
| Five packets declared, two identifiers each | ~96 Hz per packet — 960 samples/s over 10 parameters |
| Seven packets declared, two identifiers each | ~45 Hz per packet — 630 samples/s over 14 parameters |
| Parameters this engine's variant lists | 1 179 |
Of those, answering 0x22 with a width matching a catalogue row |
331 |
| Rows dropped because the car's width disagreed with the catalogue's | 91 |
| Rows that qualified as signed; rows that came back into declared range | 33; 13 |
| Rows naming an output (suffix "Command") | 29 |
The packet figure is the one worth carrying forward. Five packets are faster per
parameter and faster in total than seven, and nothing in the design predicted that:
2 × 5 = 10 parameters at 96 Hz beats 2 × 7 = 14 at 45 Hz on aggregate throughput as well
as on resolution. The manufacturer's own tool never ran more than four packets in the
captured session, which now reads less like caution and more like somebody having
measured this before.
This has not been turned into a cap in the app, and that is a deliberate reversal of
an earlier intention. Hard-capping at five packets would reinstate the ten-parameter
ceiling that S5 exists to remove. The measurement is reported; the count still follows
the selection. Choosing between resolution and breadth is the user's, and the screen now
gives them the numbers to choose with — which is what S7 is for.
3. Two readings that were impossible
Both had the same root: a catalogue identifier carries several rows and the app was
choosing among them without asking the car.
3.1 The accelerator at 798 %
0x131F carries three rows — one byte scaled by 100/255, and two of two bytes, one of
them scaled as though it were still a byte. This car answers one byte. Picking a
two-byte row applies a scale built for a width the module does not send.
The probe already had the answer and threw it away: it read the reply and kept only the
fact that something came back. It now keeps the width, CarProfile stores it as
id:bytes pairs, and rows whose width disagrees with the car are not offered. Ninety-one
of this module's 331 answering rows fail that test.
3.2 Injector balancing rates at +0.46 ms
They run −0.256 to +0.254 ms: a signed byte times 0.002. Read unsigned, cylinders 3
and 4 reported +0.460 and +0.450 where the car is trimming −0.052 and −0.062 — a healthy
engine reported as a broken one, on the screen somebody uses to judge injectors.
The rule is narrow on purpose: a declared minimum below zero and no offset in the
formula, so a coolant temperature at (X(0)*1)-40 is untouched. Across this car's
answering identifiers, 33 rows qualify, 13 come back into their declared range, and none
leaves it.
4. The drive that came home empty
A twenty-minute drive recorded nothing, because the phone locked. Three things had to be
true and none were:
- a foreground service — the ELM327 bridge had one; the part that writes the file had
nothing; - a Wi-Fi lock — Android powers the radio down with the screen, and the adapter's
access point is an ordinary network as far as the phone is concerned
(WIFI_MODE_FULL_LOW_LATENCYon Q+,WIFI_MODE_FULL_HIGH_PERFbelow); - a partial wake lock, so the reading thread keeps running.
LiveService holds all three for exactly as long as a run lasts, including when a run
ends by itself.
Three further costs the same session exposed:
- A viewer that could not open its own files. The recorder grew an
identifier
column; the reader still counted to four. On a five-column file the fourth field is the
unit,"%".toDoubleOrNull()is null, every line was skipped, and a complete
185 000-row recording opened as "no readings in this file". Columns now come from the
header. - Many files, most of them ten bytes. A sink was created per run whether or not
anything was ever written to it. Creation is lazy now: no first row, no file. - A gzip sync-flush per row. Correct when a row was a polled reading four times a
second; ruinous at seven hundred a second, on the one thread that has to be back
reading the socket before the module sends the next packet. Every 250 ms now — at most
a quarter second lost to a cut session, which is the trade the code already claimed to
be making.
5. The ten-parameter ceiling was ours
Forty-seven parameters selected; the first ten reading, the other thirty-seven never
receiving a sample.
Five packets × two identifiers. Those are limits on a message — five packet numbers is
what one AA 04 frame carries — and they had been read as limits on a plan. The
protocol numbers seven packets (0xF8–0xFE), and a start now goes out as as many AA 04
commands as it takes. Past fourteen identifiers a packet only holds seven bytes, so seven
of them hold forty-nine bytes however they are declared; the remainder is polled one per
turn round the stream loop instead of being dropped. Everything selected gets data — the
packets fast, the rest a few times a second — and the screen says which.
A related fault, found the same evening: streaming stopped after two or three seconds.
The module stops emitting about five seconds after the last 3E, and the stream loop sent
none. The bench had known this for weeks; the app had never been told. The loop now runs
inside a whileStreaming block that keeps the heartbeat going.
6. Four filter ranges instead of the whole bus
The adapter's own light was the clue: it blinks hard the moment this app connects and asks
for nothing, where the manufacturer's software sits quiet.
It was doing what it was told. The opening this app replays is the manufacturer's driver
byte for byte, and its channel message carries one filter: PASS mask=0 pattern=0 — send
everything. On this car that is about 1 340 frames a second, of which some 173 have the
exact shape of a complete ISO-TP single frame and are nothing of the sort. All of it
crossed the Wi-Fi, was parsed, and was thrown away in software. The code said so, in three
separate comments, each ending "until the app installs filters of its own".
The filters were built and proven at the car on 14 September — four PASS ranges
(0x7F8/0x7E8, 0x7E0/0x640, 0x7E0/0x540, 0x7FF/0x5E8), 541 data bytes signed with h4
and h8, accepted with a VIN reading whole through it — and had simply never been carried
across. They are now, sent after the opening as a change to a channel that is already up,
which is exactly how they were proven. A refusal is not fatal: an unfiltered bus is the
one this app has always used, and the Link screen says which it got.
7. Catalogues that ship a profile, and say when they change
Choosing what to watch is the slow part of using this. A module offers hundreds of rows,
and the dozen worth watching together are always the same dozen.
Profiles. A catalogue publishes profiles.txt — the same plain format the app exports,
several one after another, # for comments — and installing brings it along as one more
optional file. opel-vauxhall now ships one: 29 parameters for an Astra J A17DTJ,
generated by the bench from a probe of the actual vehicle, covering rail pressure demanded
against real, the four injector balancing rates, EGR commanded against position against
desired, swirl-flap position and target, DPF differential pressure and soot, both exhaust
gas temperatures, turbo vane command, MAP, MAF, IAT, torque, injection quantity per
stroke, and the pedals.
It is plain text rather than something cleverer for a reason: a row key describes its row
in full — name, identifier, width, formula, unit — because a catalogue lists several rows
per identifier and only one belongs to a particular car. So a profile is readable,
checkable and editable by hand, and can be written by something that is not this app.
Which is how this one was.
Revisions. Refresh used to re-read the source's index and compare it against nothing,
so a catalogue that had changed looked identical to one that had not. The index now
carries a content hash per catalogue, written into brands.txt and into each
plugin.json; the app compares the published one against the installed copy's own and
says update available, with the button turned into a reinstall. Unknown is not the same
as current — a catalogue installed before revisions existed has nothing to compare, and
says nothing rather than claiming to be up to date.
And where profiles actually live. The first cut put them behind the Live screen's
group dialog, which is reachable only after a scan, which needs the car. A catalogue could
ship a profile and its owner had no way to see it from the house. They are now listed on
the catalogues screen, beside the rest of a catalogue's contents, with the module and row
count and an Import that reports itself.
8. Actuators, and the wall behind them
The Settings switch that unlocks actuation had been wired to nothing since it was written:
it asked for the device credential, it flipped, and no other line in the app read it. It
now gates something, and the thing it gates is listed in front of it.
The list. A row is an output when its name ends in Command — the marque's own
convention. Anything looser fails here: "Control" takes in Cruise Control Cancel Switch,
a driver's switch; "Test" takes in HO2S Test Status, a result; "Learn" takes in
Immobilizer Password Learn, a routine on another service. Of the 331 rows this engine
answers with its own width, a loose rule matches 68 and the suffix matches 29.
Several share a byte — 0x1122 carries the three cooling-fan relays and the 2+3 pair,
0x1125 the reverse inhibit solenoid and the skip shift — so they are bits of one status
byte rather than separate identifiers.
The scoreboard, which turned out to be the better find. Three device-control status
registers answer on this ECU:
0x32EB— the control register: Device Control Requested While Other Device Controls
Active, Function Aborted, Learn Procedure Already Performed This Ignition Cycle,
Turn the Ignition Off and Restart the Engine.0x32EC— why a request was refused: coolant, engine speed, rail pressure, fuel
temperature, intake air or transmission fluid out of range.0x326D— nineteen bits of regeneration interlock: Regeneration is Not Allowed, Soot
Mass Too Large, Gear Engaged or Clutch Pressed, Accelerator Pedal Pressed, and
control lost because the tool stopped talking — the readable face of the five-second
timeout the heartbeat exists to avoid.
Their existence is the evidence that this module implements device control at all. The
capture never shows it; the ECU keeps the status words regardless.
The wall. This car is GMLAN, not UDS, so the actuator service is $AE DeviceControl
rather than 0x2F — and $AE takes a CPID, from a namespace that is not the parameter
namespace. 0x32A8 reads what the ECU is commanding the EGR to do and says nothing about
which CPID would let a tool command it. The captured factory session holds 480 requests and
not one $AE. This module's CPID table has never been observed, and inventing one means
sending invented bytes to a running engine.
So Diagnostics permits exactly one service it did not permit before, and only while the
lock is open. 0x3B, 0x27, 0x28, 0xA5, 0x34 and 0x36 stay refused either way: a
lock is the wrong control for a service that writes to a module or leaves it needing to be
programmed out of a state. The one command implemented is AE 00, which the protocol
defines rather than the module — hand every device back. It is also sent on disconnect, and
the lock closes with the link, because a permission granted for one car is not a permission
for the next one.
9. What this says about the car itself
Incidental to the release, but it is what the tool was built for, and all three were
settled by measurement rather than by looking underneath:
- DPF: gutted. Differential pressure sits at essentially nothing across the rev range
where a loaded filter would show a rising delta. The 97 % soot reading is real and
matches GDS2 — the catalogue's declared 0..1 range for that row is what is wrong. - EGR: blanked. Commanded position and actual position track each other cleanly, which
is the electronics working; the mass-flow response to those movements is not there. - SCV (rail pressure regulator): healthy. Demanded and actual rail pressure track
within tolerance from idle through 3 000 rpm. - Swirl flaps: fitted, and moving.
Intake Manifold Runner Control Valve Positionand
its desired counterpart both answer and both move. Palometas and mariposa de admisión
are not the same thing — the intake throttle is a separate valve, and this car has both. - Turbo: variable geometry, confirmed.
Turbocharger Vane Position Solenoid Valve Commandanswers with a percentage that moves with load.
10. What is not verified
- The CPID table. Nothing about
$AEon this module has been observed. There is a safe
way to find out, written up in83-actuadores.mjs: an unknown CPID is refused with
0x7F AE 31(requestOutOfRange) and a known one with the wrong length with0x7F AE 13,
so sweeping0x00–0xFFwith more control bytes than any CPID accepts separates the
two without any request ever being a complete command. Engine off, ignition on, one
session at the car. - Multi-frame
2Cdeclaration. One cheap test:2c f8 00 0c 00 05 00 43as a
multi-frame request.6c f8means the module accepts more than two identifiers per
packet;7Fmeans it does not. Untried. - Catalogue min/max. Wrong for roughly 72 % of rows in
opel-vauxhall— 34 % declare
0..1and 39 % declare0..0. This is an extraction fault in the plugins repository,
not a reading fault in the app, and it has deliberately not been guessed at: it is what
made the DPF soot reading look wrong when it was right. 0x12B1. Named Throttle Position in this variant rather than Intake Air Flow
Valve Position. The bench status line still uses the older name.- The bridge from a second device. The ELM327 bridge is self-tested but has not been
driven by a third-party OBD app over Wi-Fi. WifiLink.reachableAsIs(). Still defined, still called from nowhere.
11. Building
./gradlew :android:assembleRelease # APK, arm64-v8a and x86_64
cargo test --manifest-path core/Cargo.toml
node tools/check-braces.mjs && node tools/check-resources.mjs
The release APK is signed with the project key (OU=Astrea, CN=Varuna); a debug build will
not install over it.
12. Licence
GPL-3.0. The app contains no manufacturer data: the catalogues live in a separate plugin
repository, and the application only knows how to read them. Neither this document nor the
public repository names a vehicle identification number or an adapter serial.