Working With Diagnostic Data Cable & Software

[katflap]

Originally Posted by muzo_31

Hi folks.
I've seen a video of a russian guy running openDiagFree on a MT05.
It was set in KWP2000 mode.
Unfortunately, I cannot use it as button texts are not translated into non russian language.

I have had a play with openDiagFree, see here

http://www.chinariders.net/showpost....7&postcount=35

Though It does connect it dosen't recognise the PIDs . It is still a useful as it can be used to clear current and historic fault codes. Though i can translate it I can't do anything about the PID bit so the info that it displays for the Delphi mt05 is of know use

[muzo_31]

Originally Posted by katflap

I have had a play with openDiagFree, see here

http://www.chinariders.net/showpost....7&postcount=35

Though It does connect it dosen't recognise the PIDs .

Thx a lot Kat for the link & proposal to translate. I really appreciate.

If you say this is not so good added value, and PID are not recognized, let's focus our effort on more fruitful tools.

For your information (and all other readers here), I'm running the toolset on a QUADRO. It has a 343cc engine, using MT05 too. And the tools are running fine
I dislike this scooter world as it is very confidential, and riders are more consumers than DIY guys.

cheers.

[muzo_31]

Oops I forgot to mention: we are still in the same questioning on Quadro, to watch closed loop flag in operation, because we found out that oil temperature goes very high when cruising on an highway. We suspect the engine leaving closed loop mode quite early in RPM and running too lean at high speed.
High RPM => computer in opened loop and too learn => overtemp => oil unefficient temperature => engine break

[pyoungbl]

Muzo, your analysis is interesting. In most cases the closed loop will be leaner than open loop. That's because the O2 sensor (Lambda sensor) is correcting the ECU to keep the mixture lean enough to pass all the emission tests. Switching to open loop usually results in running richer. Note that an earlier post showed that the MT05 specs call for it to go open loop under full throttle in order to get more fuel into the engine. A rich mixture burns cooler than a lean one. I wonder if there is something else going on with your engine. It may be that the engine is staying in closed loop when it should be open loop...maybe a bad sensor for something like throttle position. I'm just tossing out ideas base on what little I know about fuel injection...not a lot.

[katflap]

Originally Posted by muzo_31

Oops I forgot to mention: we are still in the same questioning on Quadro, to watch closed loop flag in operation, because we found out that oil temperature goes very high when cruising on an highway. We suspect the engine leaving closed loop mode quite early in RPM and running too lean at high speed.
High RPM => computer in opened loop and too learn => overtemp => oil unefficient temperature => engine break

On Sunday I hope to make a PCHUD recording focusing on the closed loop indicator.

If your theory proves correct and you haven't got any bad sensors, as suggested by pyoungbl, then you are still left with the problem of how to fix your high oil temperature.

A new "map" for the ECU, I think, could be hard to find and do. Perhaps fitting an oil cooler some how may be the answer.

P.S Your scooter looks like a fun ride !

[katflap]

Originally Posted by pyoungbl

Muzo, your analysis is interesting. In most cases the closed loop will be leaner than open loop. That's because the O2 sensor (Lambda sensor) is correcting the ECU to keep the mixture lean enough to pass all the emission tests. Switching to open loop usually results in running richer. Note that an earlier post showed that the MT05 specs call for it to go open loop under full throttle in order to get more fuel into the engine. A rich mixture burns cooler than a lean one. I wonder if there is something else going on with your engine. It may be that the engine is staying in closed loop when it should be open loop...maybe a bad sensor for something like throttle position. I'm just tossing out ideas base on what little I know about fuel injection...not a lot.

I agree,

If muzo_31 problem is open loop/closed loop related,

then its more likely to be stuck in closed loop with the lean fuel ratio as you have stated .

[muzo_31]

Originally Posted by katflap

If your theory proves correct and you haven't got any bad sensors, as suggested by pyoungbl, then you are still left with the problem of how to fix your high oil temperature.

A new "map" for the ECU, I think, could be hard to find and do. Perhaps fitting an oil cooler some how may be the answer.

I don't want to spam your topic with my troubles, but we have broken 4 different machines in a few weeks time window. I would reject the sensor issue unless we have had the exact same sensor failure and almost simultaneously.
What was common fact was engine mileage. They all got more than 30'000km - 17'000 mi, and less than 40'000km - 25'000 mi. I suspect an engine design half completed, or map issue.

Anyway, I was already in thoughts related to oil cooling . My main issue to go through is to find a suitable sandwitch plate, as the oil filter is quite small. then add a dedicated radiator..
My last jocker would be to get a plate machined, but would be more conveniant to get one off the shelf (& share a tuto with virtual friends ) definitely oil is far too hot as it was measured 150°C after a 10minutes cruise at 80mph

[katflap]

I have been asked via PM to create a more straight forward post on getting PCHUD up and running. This is my attempt to do just that.

as always - AT YOUR OWN RISK

PCHUD, at present, will only run on a windows 32bit based computer.

Getting connected -
--------------------
A data cable is required to connect to your Windows laptop. There are 2 options available :-

1/ A "plug n play" cable from China - see here - http://www.chinariders.net/showpost....74&postcount=1

2/ make your own for probably under $20

Making your own cable -( there are many ways, but this is possibly the easiest)
-----------------------------------------------------------------------

components -
1/ KKL VAG-COM 409.1 USB CABLE eg http://www.ebay.co.uk/itm/USB-Cable-...-/261527808287

2/ OBD2 FEMALE EXTENSION CONNECTOR eg http://www.ebay.co.uk/itm/OBD2-OBD-I...-/281988893836

3/ three 2.2mm spade connectors ( or anything thats a reasonable fit,
I guess you could just jam in the wires if you wanted)

Using a multimeter, test for continuity to determine and isolate the 3 wires required.
crimp or solder a connector to each of the 3 wires
make a note so you can easily identify each connector ( 12v, k-line and ground)



Installing drivers for usb cable-
-----------------------------------
leave the USB cable disconnected
Run driver installer (included in download)
plug in USB cable
allow windows to do any further installation that it requires
once done, locate the installed com port in device manager. This is located in windows control panel
(cable from China - silicon labs cp210x usb to uart bridge)
(Vag com cable - USB SERIAL CH340 )

check to see that has been allocated a port ranging from 1 to 4, if not, then change it in the com port advanced settings.
make a note of the com port .


PCHUD location- (pchud included in download)
----------------
copy the PCHUD folder onto the root of your "c" drive. Inside the PCHUD folder is an executable (HUD.EXE) this is used for launching the PCHUD software.

Connecting to the diagnostic socket
------------------------------------
with the ignition off, connect to the diagnostic socket as below in the picture
(double check 12v & ground on diag. socket with multimeter)



Configuring PCHUD
--------------------
Launch PCHUD (hud.exe)
select the "setup" tab in the PCHUD window,
select the "parameter file" tab
choose a .par file (eg k_flap.par)

select the "file" tab the PCHUD window,
select the "open" tab
choose .hud file (eg k_flap.hud)
go back to ,
select the "setup" tab in the PCHUD window,
select "com protocol" tab
place bullet point in "keyword2000" and in the "device code :" box needs to read 17
ok


go back to ,
select the "setup" tab in the PCHUD window,
select "com setup" tab
select the com port that your cable is connected to
make sure "Baud Rate" is 10400
make sure "DTR high at startup" is NOT selected
ok
Switch on ignition and if all is well data will be displayed within aprox. 30 seconds

download - http://www36.zippyshare.com/v/2mLtFEuY/file.html

[muzo_31]

Thanx the kat, for your tuto !
It matches what I captured from posts and videos around the world (not so many unfortunately). I made a lower cost mod as I dedicate my VAGCOM probe to the bike only => I added some piece of wire inside the probe and drilled a small hole to let them going out. On bike side, I added a 3 terminals female connector inside the glove box, and connected it to the stock diag connector.


where
red: 12V,
green: gnd,
purple: K line

Is your plan to reach the ultimate level (= be able to setup your own maps) and load them into the computer ?

[katflap]

Originally Posted by muzo_31

Thanx the kat, for your tuto !
It matches what I captured from posts and videos around the world (not so many unfortunately). I made a lower cost mod as I dedicate my VAGCOM probe to the bike only => I added some piece of wire inside the probe and drilled a small hole to let them going out. On bike side, I added a 3 terminals female connector inside the glove box, and connected it to the stock diag connector.

Nice work muzo, and even cheaper

Originally Posted by muzo_31

Is your plan to reach the ultimate level (= be able to setup your own maps) and load them into the computer ?

No, not really, just to learn about and gain EFI diagnostic skills, the map for the RX3 seems to work really well

[katflap]

Originally Posted by katflap

On Sunday I hope to make a PCHUD recording focusing on the closed loop indicator.

Here is the (ply) recording.
http://www28.zippyshare.com/v/0PgGDr6q/file.html

Initial observations
------------------------
AFTER INITIAL WARM UP

1/ C L enabled is iluminated all the time

2/ The short term fuel trims (CL STFT COUNT, CL FUEL CORR FACT ) seem active all the time except when the :-

ACCELERATION ENRICHMENT ENABLED, (ACC ENRICH ENABLED )
DECCELERATION ENLEANMENT ENABLED (DECC ENLEAN ENABLED)

( though its dificult to determine if the ECU is acting on this info.)

3/ The AFR TARGET and the AFR COMMAND remain at 14.6 unless the throttle position is 75% or more , when the air fuel ratio gets richer.


My initial impression
----------------------
It seems that the ECU does try to keep in closed loop operation as much as
possible ( AFR 14.6) and only breaks to open loop on hard acceleration/ deceleration or getting close to WOT.


Points of note in ply recording
------------------------------
1/ shortly after 00.10.55.560, I try and hold a steady 4500rpm, approx. 40mph (indicated speed)
2/ shortly after 00.12.14.106, set off down the highway
3/ shortly after 00.12.26.384, see AFR change from 14.6 to 13.15
4/ shortly after 00.12.29.458 I try and hold a steady 7000rpm, approx. 70mph (indicated speed)

Unfortunately the recording cuts at 13 mins

If any one who wants to see this but is stuck with windows 64bit computer, try my PCHUD emulator.
http://www.chinariders.net/showpost....&postcount=106

Place the sun2.ply file here :-
C > PCHUD64 > DOSWIN > PCHUD > PLY

[pyoungbl]

katflap, that's great information. Since our O2 sensor costs something like $130 I'm starting to wonder if it is actually wide band vs the narrow band you would expect in a budget system. Wide band would let the ECU stay on or close to the target AF ratio under more conditions. Your thoughts?

Peter Y.

[katflap]

Hi peter,

my best guess is that we are using the narrowband O2 sensor.

Apparently a wideband sensor has 5 or more wires, ours only has 4.

Not a very technical answer but all I have at the moment.

I didn't realize they were so expensive

I had another look over the recording, I decided to ignore the "AFR TARGET" &" AFR COMMAND" readings and focus on the "O2 RICH" indicator and the O2 voltage".
Assuming that an O2 voltage of 450mv is equivalent to an AFR of 14.6:1

and indeed it did appear to be running at this AFR for most of my trip, with the "O2 RICH" indicator alternating between rich and lean equally. Convincing me more that the EFI is held in closed loop more than originally thought.

It was also clear that when it entered "POWER ENRICHMENT MODE" at about 75% throttle opening, it was definitely in open loop.

[muzo_31]

I have a wideband oxygen sensor for measurement purpose only (used to set carb jet size). it can provide a voltage from near zero V up to near 5V.
If your probe can generate a few tens of mV around 0.6V, it is a narrow band one = lower cost. (and I believe so, for mass prod' bikes)

[katflap]

Thanks muzo,

Ours is narrow band then, it only does 0v - 1v