[pitrack_1]

The "more later" bit on the white car (diesel)

Been a bit distracted so it's taken me the week to get to this.

The Forester was in and the dealer kept it overnight. They seemed to do a bit of test driving (judging by the klms), no doubt logging and in the end they 'decarbonised' the engine by squirting Subaru Upper Engine Cleaner in through the glow plug ports.

I believe it's a bit of a job to get to those glow plugs?

At any rate, the vehicle has been returned after all this, out of the warranty period, at no cost to me (so far), except for a bit of diesel. So, in my case, well done to Subaru and the dealer- where the Service Manager has pretty well overseen the handling himself.

The result? The car drives OK...general performance seems unchanged although there *may* be some economy increase. There do seem to be slightly fewer regens, they are smoother and they are now harder to pick in general driving esp. with cruise control. However, they do still cause performance lag and can be quite noticeable on certain occasions. I'm beginning to believe that the performance varies as the DPF regen itself varies.

Some graphs are attached highlighting my CBR-SYD and SYD-CBR trips last weekend. Trips are similar except for the weather being mild to very warm and humid, with some rain. Temps varied between about 20C and 30C. As you can see, excluding regens (based on commanded throttle actuator) at the beginning of logging, one trip had 2 regens and the other 3.

I'll have to drive it more to evaluate further.

 

[Amboscott]

So why is it regening that often in what should be passive regen conditions?

 

[pitrack_1]

More and less...

I'll have to drive it more to evaluate further.

Lots of Canberra based driving this weekend. Regen every 60 kms or so. One left me stranded in a merging manoeuvre, lucky the other vehicle didn't insist on maintaining his speed (all of 60km/h).

So why is it regening that often in what should be passive regen conditions?

Perhaps: If it's in (passive) regen conditions, why is it initiating a (active) regen at all?

 

[dpm]

Definitely not right. Even at my high miles I still only see a regen every 200+ miles. Can you log exhaust pressure differential or either exhaust temperature?

Either the filter is being blocked by poor combustion, it thinks it is due to a dodgy sensor, or the ECU is getting it's miles/time between regens calculation wrong.

 

[pitrack_1]

The "more later" bit on the blue car (Impreza loaner)

Here's a quick opinion/comparison between and old(sh) apple and a new orange, although the Subarus parts commonality makes it reasonably acceptable...

Impreza Petrol CVT vs 2010 Forester Diesel

Summary:A short-term loan of the (base?) Impreza was enough to put it on my evaluation list for a Mazda-3-type replacement. Composed, quality behaviour, much improved quality interior, acceptable performance. And the 3-y.o. liked it!

Interior: After being scared off by the previous model Impreza which was even more downmarket than the Forester Diesel, this Impreza was a marked improvement. Better quality plastics, esp. finish-wise, better displays with even lighting (except the real el-cheapo LCD in the dash...what's with that?). Stereo better than Forester. Good rear boot for a hatch. Seating OK. The climate control behaves in the Impreza, unlike the Forester. Impreza's stereo seems better (still not great), centre display top-notch to read, switchgear better but steering-wheel-mounted controls can be fiddly and buttons very cheap (worse than Forester's) and no mute button for the stereo. No temp gauge and fuel 'gauge' a damn hard-to-read bar on the dodgy LCD screen. No courtesy light on map lights made interior really dim (and may not have had a time delay, either- can't remember). Fit and finish looked typically good apart from the dodgy steering wheel buttons (esp cruise).

Steering, Handling, Roadholding: Not tested to any great extent, but the Impreza pushed into understeer quite early, which took away from the experience and my confidence. Smaller wheels/tyres may have played a part. Apart from that, it was far more refined, damped, quiet, flat and generally composed that the Forester, which was a noisy, bouncy, roly-poly, rough, utilitarian thing afterwards. But the Forester remains neutral, probably due to the drivetrain drive split. The Impreza's understeer prevalent in cornering (compared to the Forester) I'll attribute in part to the different 4WD system and slightly different drive split (60:40 Impreza, 50:50 Forester). The Forester remains more neutral in corners, esp under (some) acceleration. But the Forester will be bumped off line, roll and jiggle whereas the Impreza remains quiet and well composed. I need to try a manual Impreza.

Engine and transmission/drivetrain: The CVT takes some getting used to but I can see how it will replace automatics. Idiosyncrasies include whine (esp. low speeds), unusual (compared to std. autos) shift patterns, holding constant revs, and noise/drone resultant from these. I encountered some tactile anomalies that need rectification, probably a result of the CVT trying to be 'intelligent' and guess my driving pattern: In the city the response is overly touchy to zero or very minor throttle- the type used gliding along at 20-40km/h. Sink the boot once gliding (e.g acceleration to merge, into a gap, to overtake, or behind another vehicle at the lights) however, and it all collapses as the CVT and engine try to reconfigure themselves to do something meaningful. Large/full throttle take-offs provide much better response (front vehicle at the lights). The 'intelligence' seems to kick in when you get a run at things (e.g. corners) and use the throttle/brake several times- it all becomes more responsive, but that's not consistently possible in city driving. I think basically we (the CVT intelligence and myself) ended up fighting each other until we arrived at an uneasy truce broken by occasional throttle-provoked skirmishes. Oh, you also get shifting paddles if you wish to play that game. A real handbrake still prevails. A Diesel CVT Forester could be a seriously good combination.

Fuel Economy: Driven gently, Impreza fuel economy is reasonable with approx 8l/100km reported in gentle city driving. Forester: ~7-7.5l/100km measured (8+ when my 'better half' gets at it). The Impreza has an auto stop-start: Now this one's weird. Unless deactivated, it will stop the engine when you hold the brake on at lights, but only for 30 secs or so. the A/C will restart the engine, but so will something else, as it restarted with the HVAC turned off. The slight delay restarting exacerbates CVT delays at attempted quick take-offs and it will shut the engine down (to immediately restart it) as you attempt a parking manoeuvre. Worst is the engine stops when parked, restarts as you apply the park brake and remove your foot from the foot brake, only for you to immediately shut the engine off with the key! None of these shenanigans happened in a VW Passat I drove overseas. Easy fix: turn it off.

 

[pitrack_1]

Logs and interfaces

Definitely not right. Even at my high miles I still only see a regen every 200+ miles. Can you log exhaust pressure differential or either exhaust temperature?

Either the filter is being blocked by poor combustion, it thinks it is due to a dodgy sensor, or the ECU is getting it's miles/time between regens calculation wrong.

Will have to invest in some extra hardware. My VAG-COM KKL cable has the 'wrong' chip in it- won't talk, physically checked the chip too.

Will probably get another cheapo USB cable (with correct chip: FT232BL). Plus perhaps a better interface- OBDKey (with BT) looks the go as it claims to do both OBD-II and SSM (just $$$).

 

[pitrack_1]

Update: My latest tinkering and fiddling- what's done and what's to come

Well, after a hiatus ('cos someone was given a clean car and brought it back filthy..grrrr! :icon_evil from driving the thing, Easter and ANZAC day coming with the time off in between (meaning: some long trips) have provided me with an opportunity to get stuck in.

It's also helped to let the 'dust' in my mind settle a little. Now I have a better grasp of what's going on and should go on. I'll post more updates including details and screenshots in the next week or two, so stay tuned (sorry, another bad car pun) as I endeavour to work this in whilst actually trying to have a family holiday! Deep breath, here goes...

Done/Achieved

• Hit a limit with my PC software, so switched to Torque Pro on Android, as per Ghostgum earlier (and elsewhere).
• Able to get the Extended/European OBD (E-OBD) parameters relating to DPF regen/oil dilution all running qualitatively, if not all quantitatively.
• Confirmed I could read out the DPF Active Regeneration Switch, showing active, during a regen.
• Configured the Parameters under Torque Pro for display and logging.
• My replacement/new VAG-COM-type cable, purportedly with the proper serial chip, has arrived!

Of course, the vehicle has sensed it's under surveillance and I'm doing something constructive, useful, important wink and likely to catch it out, so it has steadfastly refused to misbehave. However a regen isn't far away now (60% soot, 65% should be the kick-in) so it's a matter of time.

To Do/To Come
I think what's coming may actually produce a tangible, useful output. Which would be a contrast to all my (mostly) hot air that's gone before...

• Snapshots of what I've done.
• Descriptions of what I've done, on what I've done it, how to do it, with what equipment and what $$$ it may cost.
• In particular, I hope to show how to set up a DPF regen warning device/lamp/light (plus other assorted info).
• Perhaps show the Subaru dealer. :lildevil:
• Fix some of the quantification issues. In particular, one of the parameters returns two bytes, not one so I need to do a bit of testing so it scales more-or-less properly.
• Test my new VAG_COM cable with FreeSSM.

And just to whet your appetite see the attached image, which was taken just after a cold start.

 

[orienteer]

Look forward to hearing more! Especially the new VAG cable, as I've never been able to get mine to work with FreeSSM.

 

[pitrack_1]

It can be done! But...

...firstly:

Do not try to control, manipulate or change the logging device whilst driving. Get someone else licensed to do the driving for you.​

Secondly: and this is my long-held based-on-experience paranoia as a techie...Performance of any useful action on a system will be actively opposed by the system. I reckon it actually comes back to the Second Law of Thermodynamics, basically that the entropy (disorder) of an isolated system always tends to maximum. So by trying to measure, control and manage the system I am actually opposing the 2nd Law of Thermodynamics, therefore the system tries to stop me...Feed me the medication now!

To wit: Since I figured out what to do, with what to do it, how to do it, when to do it and actually did it,

• I've had 2 regens when before I had (and logged) 8-10 regens.
• Required another driver- my wife (i.e. doubling the waste of person time).
• Had my screenshots (taken by simultaneously pressing the 'Volume down' and 'Power' buttons) destroyed by a perfectly placed volume bar which didn't seem to be there
  • when I took the screenshot
  • and on the icon representation of what was taken.
• My pre-regen warning alarm has already been falsely triggered by the Soot level wandering up there and then going back down again.
• Note the image posted yesterday of the glow relay wasn't harmed.

I apologise for the screenshot vandalism below. A soon as I can get the *&^% vehicle and $#@! tablet to comply I'll get proper screen shots posted.

Details

• The display setup is mine, so I can watch and get an idea of how things change.
• Parameters are also logged for future plotting/analysis.
• Note the EGT@DPF value is incorrect, although the qualitative behaviour is correct. I need to rescale the calculating equation, I'm probably only ~20% out.
• DPF regen starts at a Soot level of 65%, driving/vehicle conditions permitting and proceeds for approx 12.5-13mins.

I'll post more info later about what parameters to set up and how.

Basic screen setup:

Just after a cold start:

Just after DPF regen initiation. Note the temps.

During DPF regen: Note the temps and soot level now.

Immediately after the end of DPF Regen. I was literally trying to press the screen capture buttons when it turned off. Note the temps and soot level again.

I've also set an alarm to warn me before the regen starts. Note this isn't foolproof (as has already been shown)

More to come.

 

[dpm]

See if the soot ever comes down lower than that 32%.

 

[gwk]

Details

• The display setup is mine, so I can watch and get an idea of how things change.

When you say this, do you mean you've put it together from the available options in Torque Pro?

If I was using Torque Pro could I put the same display together?

 

[pitrack_1]

Torque Pro displays

When you say this, do you mean you've put it together from the available options in Torque Pro?

Qualified Yes. You need to enter and configure some custom PIDs.

If I was using Torque Pro could I put the same display together?

Yes.

Let me type the details of what to do/how to do it up in some semblance of order before I post it. BUT...I don't know how (or whether) it will work with the MY13. The engine 'internals' changed substantially for MY12 (I think) and it's quite possible all the computer stuff has changed along with it.

 

[pitrack_1]

DPF soot values and behaviour

See if the soot ever comes down lower than that 32%.

Ahhh, dpm, you've smelt a rat all the way along- I just haven't been able to present it to you!

The soot does decrease below 32%, all the way to 0%. But how and when it gets there is the interesting bit.

That 32% was driven (sorry again!) by a regen in city conditions. Only the last minute of that regen had freeway (80km/h plus) conditions, so that 32% was like 45-52% a minute or so earlier.

Very quickly after being able to watch the soot level I started to suspect the cause of my problem is the soot level measurement, effectively what I think you've suggested all along. This is because the soot ratio behaves in rather inconsistent and somewhat unpredicatable manners. I do not believe the reading is reliable, at least in my case.

It generally likes sitting in the region of ~52-64%
The best way to explain is probably to outline what conditions relate to either decreased or increased soot levels, at what rate, and by how much.

Increased soot levels

• Acceleration, e.g. getting up to speed from traffic lights. Usually adds 1%
• General increase during urban driving
• When engine's cold. This is often a quick and substantial increase
• When coasting down. This can be as quick as 1%/sec and add 10-15% for a coast down from freeway speeds. Note the l/100km is reading 2.0l/100km (min possible readout) during this and no regen is taking place.
• The DPF temp being somewhere about ~190C?
• A low soot level tends to rapidly rise

Stable soot levels

• A soot level between ~50% and ~64%. The soot value seemingly can stall in this range for 100's of klms.
• 110km/h on the freeway.

Decreased soot levels

• Harder driving, e.g. ascending a climb
• Sometimes, just freeway driving (esp undulating)
• Engine braking with the revs up whilst descending a steepish mountain pass. This drove the soot value down to 0% quite rapidly. Now why would that be when other coasting increases it ?!!
• Regenerations

Now if my logger had logged today, I would have been able to show all this...

 

[pitrack_1]

Ha! Finally!

And here's the proof you can monitor a DPF regen using an ELM-327-type OBD-II adaptor and an Android device with Torque-Pro...

The regen...note the temps!

The DPF regen imminent warning...which is a really bad idea, at least until we can get some control e.g. of hysteresis levels, time outs, silencing, etc. Note this is the actual DPF regen warning. The DPF imminent warning is similar but triggers off a 63% soot level.

 

[dpm]

what custom PIDs do you have? I've always fancied setting up X-gauges on the Scangauge but have never quite got round to it.

Try looking at deltaP and seeing how it matches sootload.

 

[pitrack_1]

The “DPF Regen On” light via OBD-II/EOBD on an Aus spec 2010 Subaru Forester Diesel

Introduction
For those ‘in the know’, here’s the bare bones you need to get the “DPF Regen on” warning light I set up on my display in the posts above. I’ll assume you already have

• Torque Pro set up on a Bluetooth-capable Android device
• A paired Bluetooth EOBD/OBD-II interface that Torque Pro can communicate with (a properly functioning generic ELM v1.5 interface that can access the CANbus works for me)
• Familiarity with the hardware/software

For those ‘in the know’, I’ll put in my next post the ‘full caboodle’ of PIDs, parameters, option, etc. I used.

For the more uninitiated, I’ll try to write some more step-by-step instructions for another post.

Thanks
Thanks must go to those who found and made this info available. Others discovered it, I just managed to make it display in Torque Pro. In particular, people/ sites such as Subdiesel /Subaru Diesel Crew (subdiesel.wordpress.com) and Ghostgum at ausubaru.com. And any others I have forgotten to mention.

Custom PID value: 22 12 5B
In “Settings -> Manage extra PIDs/Sensors”, add a custom PID with the following values, without the inverted commas, my comments are in ():

• OBD2 mode and PID: “22 12 5B” (Mode: 22, PID: 12 5B)
• Long Name: “DPF Active Regeneration Switch”
• Short name: “DPF Regen”
• Min value: “0”
• Max value: “255” (equates to FF in hex)
• Scale factor: “x1”
• Unit type: “on/off” (this may be non-critical)
• Equation: “A”
• Leave the OBD Header, Diagnostic Start and Diagnostic Stop fields blank.

If you’re connected to your OBD-II device, you can click “Test”. The two values that should be returned are either “00” for OFF (no regen) and “FF” for ON (regen in progress). Click OK when done.

DPF Regen Display Light Display
Go to one of the real time display screens, and from the Menu:

• Select “Add display”
• Select “On/Off Display” (down near the bottom of the list for me, shows “on” and glows green if value is >0)
• Select “DPF Active Regeneration Switch” from the PID list
• Select the Size of display you want

And Off You Go!
The light will glow green and say “On" when the DPF regen is in progress and Torque Pro is linked to the car through the OBD-II interface.

 

[pitrack_1]

The Extended PIDs for DPF monitoring...finally!

Well, here they are below. The extended PIDs I have for monitoring various diesel parameters.

I've also attached Torque Pro files for both the extended PIDs and the dash I use. These should prove helpful for those wishing to import the info, instructions are also below.

Again thanks to those who posted PID info previously, in particular ghostgum at ausubaru and Subdiesel's Subaru Diesel Crew site.

Enjoy! And Don't fiddle whilst driving!

Instructions for importing the files (once you've downloaded them).
Both files are text-readable so have a squizz and see the layout. The PID file in particular should be easy to modify.

• If you have any current custom PIDs or dashboards you want to keep, back them up now
• The extended PID file is called "Forester_DPF_Monitoring.csv" and in csv format. Place it in Torque Pro's "extendedpids" subdirectory (my torque base directory is at /storage/emulated/0/.torque/). Go to torque -> Settings -> Manage Extra PIDs/Sensors, open the menu and select "Add predefined set". It should be listed there, select it.
• The dash file is called "Forester_DPF_Monitoring.dash.txt". The ".txt" was added to allow the upload and should be stripped form the filename to give "Forester_DPF_Monitoring.dash". I have not tested importing it, but the file should be placed in Torque Pro's "dashboards" subdirectory (my torque base directory is at /storage/emulated/0/.torque/). Import by going to "Realtime Displays" and selecting the "settings" icon. Then from the menu "Layout Settings" -> "Import Layout" from the menu. It should be listed there, select it. I do not know if this will supplant your current dashboard or augment it.

DPF Related Custom PIDs
Notes:

• The formula (A-2)*25 + B/10 for EGT@DPF may not be quite correct, but it should be close. Qualitatively it's correct, let me know if you hit the right one.
• PID values are in hexadecimal. The other values (e.g. max, min, equation constants) are decimal.
• Is there are reason the temp formulae were not factorised, i.e. 5A - 40 is not 5(A - 8)?
• The funny characters should give the "degrees C" symbol properly. I think!

Name: Glow Relay Switch
Short Name: Glow Relay
Mode And PID: 22124F
Equation: A
Min Value: 0
Max Value: 255
Units: on/off
Header:

Name: Oil Dilution Ratio
Short Name: Oil Dilution
Mode And PID: 22114E
Equation: A
Min Value: 0
Max Value: 255
Units: %
Header:

Name: Estimated DPF Temperature
Short Name: Est. DPF Temp
Mode And PID: 22114C
Equation: 5*A-40
Min Value: -50
Max Value: 1000
Units: °C
Header:

Name: Estimated Catalyst Temperature
Short Name: Est. Cat Temp
Mode And PID: 22114B
Equation: 5*A-40
Min Value: -50
Max Value: 1000
Units: °C
Header:

Name: Soot Accumulation Ratio
Short Name: Soot
Mode And PID: 22114D
Equation: A
Min Value: 0
Max Value: 255
Units: %
Header:

Name: DPF Active Regeneration Count
Short Name: DPF Regen Count
Mode And PID: 221157
Equation: A*256+B
Min Value: 0
Max Value: 65535
Units:
Header:

Name: DPF Active Regeneration Switch
Short Name: DPF Regen
Mode And PID: 22125B
Equation: A
Min Value: 0
Max Value: 255
Units: on/off
Header:

Name: Exhaust Gas Temp at DPF
Short Name: EGT@DPF
Mode And PID: 22003E
Equation: (A-2)*25+B/10
Min Value: -50
Max Value: 1000
Units: °C
Header:

 

[pitrack_1]

Fault Code U0073

Has anyone seen this U0073 code before, and has anyone more info on it?

Well, just typical of this car. Set off for the drive from Mollymook to Bombala, and hope to take a scenic dirt road (Tantawangalo Mtn Rd). Get to Narooma and 3/4 of the lights on the dashboard pop up, the temp light flashing alternately blue/red to boot.

Retire for lunch, come back and start the car- CEL and TCS lights are on. Luckily able to download the code and then clear it (which clears the CEL/TCS lights) to see if it recurs (it hasn't so far in ~6hrs of driving). Meanwhile, retreat back to tar roads.

The code generated is U0073, no real clear definition but which seems to be a network error/communication bus error ("Control Module Communication Bus "A" Off). It has a few references in Subaru-related forums, often to do with logging.

Thanks,

 

[dpm]

Never seen it, but would guess that you were taking enough data off the CANBUS that some "important" message was somehow dropped. Shouldn't happen (as the system is designed to work on different levels and speeds and be robust) but...

 

[Turbo_sarah]

Has anyone seen this U0073 code before, and has anyone more info on it?

Never see it in my car bevor, but it mentioned:

U0073, CAN-error, bus-off detected

Seems you killed the CAN-bus temporary...