2006+ Honda Civic Forum banner
21 - 40 of 50 Posts

·
Premium Member
Joined
·
131 Posts
Discussion Starter · #21 ·
Here some update: I have been offered a "comparison" with an identical car, to hopefully pinpoint wether it is some mechanical problem or a software issue.

The results (see screenshots taken from Honda HDS (full resolution in ZIP), marked line is the point where max. boost is reached):

1) My car is obviously lacking performance when compared to the other civic (2.2 i-CTDi, MJ 2007 with DPF) - technician mentioned it is feeling dull.

2) My boost pressure is about 0.1 bar lower (e.g. i get 2.07, the other car has 2.17 or even 2.22), and drops off quicker once past 3000 RPM; also the max. boost is reached some 200 rpm higher in the other Civic -> it has noticeable better performance in the high (3000-4000) rpm range. It feels like 10-15 BHP difference.

3) Boost control solenoid duty cycle is slightly higher on mine, which should mean either the turbo has to work harder for the same amount of boost, or that more force is needed to move the VTG. So either a small leak, or a stiffer VTG (but obviously within normal limits, as the duty cycle stays far from any extreme values and changes smoothly). Perhaps a lower injection quantity -> less energetic exhaust gasses could also need a more closed VTG, to get the required boost to maintain the right air-fuel-ratio.

4) The other car had an older ECU map installed, Version "37805-RSS-E260", mine had "37805-RSS-E280". The car with E260 (the better performing) had a more aggressive throttle response, more power at high rpm, and during the first tries to get data the clutch slipped so bad that we considered aborting the test. However, the slip was gone once the clutch got warmer. I have nearly twice the mileage, and no slipping clutch.


So for now, i would conclude:

a) The lack of performance is real and backed up by comparing basic data of identical engines.

b) Wether it is based on mechanical problems or a more restrictive ECU map (clutch protection?), is still open, although inofficially, and by jugding what Honda tried to do (change ECU for free against one with an old map, which unfortunately did not work as intendend (delivered ECU had new map installed)), it might be possible that the latter could in fact be true.

The recorded data have been sent to Honda for investigation. In the meantime, i will try to get my ECU map downloaded, and have it compared to an older one. I could get an old original map from a tuner, but it would cost the same as a remap...
 

Attachments

·
Premium Member
Joined
·
131 Posts
Discussion Starter · #22 ·
I am happy to report "problem solved":

My suspicion that Honda has "detuned" my engine with a software update was actually spot on!

Thanks to the help of a tuner friend, my ECU was flashed back to its original state (the most powerful stock version available directly from Bosch) - it almost feels like a remap now.

Boost pressure went up from about 2.10 bar to 2.25 bar (@~3000 rpm), and ist maintained longer in the upper rev range (2.10 bar at 4000 rpm vs. sometimes below 2.0 bar before).

There is still a small "fluctuation" when i floor it a around 2000 rpm, although less pronounced than before, and without apparent impact on performance. The range from 2000 - 2500 rpm is now much more usable, normal overtaking maneuvers are possible in 5th or even 6th, were i had to change down to 4th before.

Indicated fuel consumption also improved by about 1 L/100 km - even with "trying out" the "new" power, i managed to get 5.3 L / 100 km, on a 130 km trip starting at 200 meters above sea level, rising to 600 meters above sea level. With the old software, the best i could ever get under similar conditions on the same route, was 5.7 L/100 km, although this required a very economical driving style. With a more "normal" style of driving, about 6.2 - 6.3 L/100 km were the norm.

I have a bunch of snapshots taken with the Honda HDS (the dealer diagnostic computer) here, which i am now going to thoroughly analyze to see what was actually changed. By just looking at the boost pressure target, which went from 2.20 bar to 2.35 bar, it is pretty obvious that there were some real BHP missing.As i normaly drive at an altitude of about 600 meters above sea level, i possibly cannot reach the full pressure, because the ECU rather seems to keep the pressure ratio constant than boost pressure (i think Relic mentioned that some time ago).

Before the "downgrade", the boost pressure spike at 2000 rpm always exceeded the boost pressure target, which forced the ECU to compensate by changing the VNT position and (possibly) cutting back injection quantity. Then, the boost dropped below the target, and the ECU struggled to get it back up again, which would take 2-3 seconds in most cases. By the time i got "acceptable" boost, the action was often almost over, leaving me with the feeling of a somewhat dull performance.

Now, the pressure spike does not exceed the boost target (which is at 2.35 bar, spike goes only to 2.20 bar), therefore there is no need for corrective action by the ECU, the target airflow is reached, and therefore the maximum specified quantity of fuel is injected.

As for now, there are no obvious issues with the downgrade, the DPF regeneration started at the expected intervall, was completed in the normal amount of time, and there is no hint of clutch slip. Throttle response is also sharper and more linear, before it often felt like nothing is happening, then the power suddenly kicked it.

I am really glad that this was finally sorted out now - after a year of not knowing what was really wrong with my car.
 

·
Premium Member
Joined
·
131 Posts
Discussion Starter · #24 ·
And here some data (excel spreadsheet) from the recordings with the HDS:

You can see that the MAP target is higher now, that the fuel injection quantity is higher, and that the MAF is higher too as a result of the increased boost.

The injection timing (not included in this file) has also been changed (e.g. at 4063 rpm, the timing was 30° before and now it it 32.5°), although i do not know wether this means later or earlier (timing starts at a couple of degrees at low rpm and get bigger with increasing rpm).

The only question that remains: Why was such an update done in the first place, and why did apparently no one ever notice it, at least not someone who is also present in the Civic related forums?
 

Attachments

·
Registered
Joined
·
4 Posts
I am happy to report "problem solved":

My suspicion that Honda has "detuned" my engine with a software update was actually spot on!

Thanks to the help of a tuner friend, my ECU was flashed back to its original state (the most powerful stock version available directly from Bosch) - it almost feels like a remap now.

Boost pressure went up from about 2.10 bar to 2.25 bar (@~3000 rpm), and ist maintained longer in the upper rev range (2.10 bar at 4000 rpm vs. sometimes below 2.0 bar before).

There is still a small "fluctuation" when i floor it a around 2000 rpm, although less pronounced than before, and without apparent impact on performance. The range from 2000 - 2500 rpm is now much more usable, normal overtaking maneuvers are possible in 5th or even 6th, were i had to change down to 4th before.

Indicated fuel consumption also improved by about 1 L/100 km - even with "trying out" the "new" power, i managed to get 5.3 L / 100 km, on a 130 km trip starting at 200 meters above sea level, rising to 600 meters above sea level. With the old software, the best i could ever get under similar conditions on the same route, was 5.7 L/100 km, although this required a very economical driving style. With a more "normal" style of driving, about 6.2 - 6.3 L/100 km were the norm.

I have a bunch of snapshots taken with the Honda HDS (the dealer diagnostic computer) here, which i am now going to thoroughly analyze to see what was actually changed. By just looking at the boost pressure target, which went from 2.20 bar to 2.35 bar, it is pretty obvious that there were some real BHP missing.As i normaly drive at an altitude of about 600 meters above sea level, i possibly cannot reach the full pressure, because the ECU rather seems to keep the pressure ratio constant than boost pressure (i think Relic mentioned that some time ago).

Before the "downgrade", the boost pressure spike at 2000 rpm always exceeded the boost pressure target, which forced the ECU to compensate by changing the VNT position and (possibly) cutting back injection quantity. Then, the boost dropped below the target, and the ECU struggled to get it back up again, which would take 2-3 seconds in most cases. By the time i got "acceptable" boost, the action was often almost over, leaving me with the feeling of a somewhat dull performance.

Now, the pressure spike does not exceed the boost target (which is at 2.35 bar, spike goes only to 2.20 bar), therefore there is no need for corrective action by the ECU, the target airflow is reached, and therefore the maximum specified quantity of fuel is injected.

As for now, there are no obvious issues with the downgrade, the DPF regeneration started at the expected intervall, was completed in the normal amount of time, and there is no hint of clutch slip. Throttle response is also sharper and more linear, before it often felt like nothing is happening, then the power suddenly kicked it.

I am really glad that this was finally sorted out now - after a year of not knowing what was really wrong with my car.
Hi mschwab,

mine is a CR-V 2.2 i-ctdi 2007 with DPF, same syntoms;
i've spent the latest years trying to figure out what's happening.

I don't have tools such HIM + HDS to get a detailed analysis like your, no chance to compare with a no-dpf CR-V too.

I recently dyno-rolled at a trusted tuning garage:
the result was max torque of 315.4 Nm shifted at 2549 RPM, max power 139.95 CV at 4144 RPM.

No clutch, MAF, MAP, fuel pressure, EGR issue at all; everything was checked;
boost, MAF, fuel pressure values are quite near the target values, however the lacks were considered negligible.

The only aspect that came out was instead the throttle behaviour.
In a portion of the throttle travel below 50% of the full range there is no change in boost, fuel pressure, air flow.
The throttle runs fine from 0% to 100% but at certain point there are flat values at every gear.
That is what i was told.

From 1500 to 2500 RPM the engine power and torque seem cutted off, more clear if the throttle is released and pressed again;
the car seems pulling back instead of driving forward and the engine revs up very smoothly and slowly;
past 2500 RPM is too late to get something of interesting.

Now, i agree with you about the detuning theory.
In addition to the clutch conservative solution, i think that a DPF-trouble-free approach was used too:
1) to get more torque at lower rpms more fuel is needed; more fuel means more unburnts in the exaust gases and more DPF clogged, so the ECU was detuned.
2) with a detuned car, to get a more quick response, driving hardly and changing gears at upper rpms (due to the max (lowered) torque shifted) are needed too; this creates the conditions and gives more chances so that the DPF cleaning cycle can easly engage.

Of course the car was completely different with the original software.

Did you dyno-tested before and after the ECU restore!?
Can i contact you via pm!?

Regards
 

·
Premium Member
Joined
·
131 Posts
Discussion Starter · #26 ·
I did not do a dyno test, but i have real word road data which indicate a 12% increase in acceleration under comparable conditions (same tires, loading, air temperature, wind, air pressure and same stretch of road).

This fits well to some dyno results here in Austria (mostly DPF cars), which are all in the 150-nearly 160 engine HP range, with torque up to 370 Nm. Also, when i can reach a speed of 211 km/h (GPS!) on a flat german Autobahn (A8 Salzburg-Munich, long straight parallel to the Chiemsee), into a light headwind, and 217 km/h without wind, this cannot be possible with "only" 140 HP.

It should be possible for you to get back your old software via a tuner, Honda will not do this.

While the downgrade of the ECU software got me back my old top speed and also improved acceleration, my problems are still not fully solved, as i had to find out when it got warmer in the last few weeks.

It still have the pressure spikes, but now with drops far below the target, down to about 1.8-1.9 bar absolute, which is enough to trigger the smoke limiter which cuts back injection quantity. Then, with reduced injection, the turbo struggles to get the boost up to the target again, and i do not really get out of the smoke limiter after it.

I cannot really explain yet why this is happening, the turbocharger actuator test with the HDS shows no problems, there are no DTCs, the air hoses were checked for boost leaks, i have tried hard running to clear the VNT-vanes, even put BG-244 in, nothing really helped.

The boost falls without the actuator duty cycle changing, like the actuator or the vanes were moving themselfes on there own. I can only think of either some sticky spot in the middle range of the VNT-vanes or actuator, which makes hitting the boost target impossible or at least hard (i can reach the target only in 6th gear at high rpm and several seconds of full throttle!), a vaccuum system leak which reduces the holding force of the actuator to a point where it runs to a low boost position, a faulty solenoid (not reacting according to commanded duty cycle in a certain range), or a strange kind of boost leak which only manifests in hot weather.

I will upload some data later, and i will check if the solenoids for EGR and the turbo are the same (according to the parts catalogue they are) and try swapping them out.
 

·
Premium Member
Joined
·
131 Posts
Discussion Starter · #27 ·
Here the recordings which document the problem. I have included one from winter, where i thought the problem was resolved, as reference.

The files are in German, here the translations:

Magnetventil Ladedruckregelung Lastfaktor = boost control solenoid duty (higher % means more boost requested)

Gesamtmenge = total fuel injection quantity (IQ)

APP = accelerator pedal position

MAP = manifold absolute pressure (boost pressure + atmospheric pressure)

MAP Ziel = manifold pressure target requested by the ECU


The curves are color coded, the x-axis is time in seconds, the y-axis shows the selected (highlighted) parameter.

As you can see in the log in 6th gear (was from 110 to to 130 km/h), the boost first overshoots the target by about 0.2 bar, then drops below target without any noticeable change of the solenoid duty cycle (e.g. going to lower to open the VNT vanes a bit). It looks like the boost is dropping on its own.

Either the VNT-actuator moves on its own without being commanded to do so (not enough vaccuum or solenoid fault), sticks at some spot so that it can only be either above or below the necessary position, or the solenoid sticks somehow in the 40% range.

The HDS boost test is completed without fault: RPM is set at about 1600, then the boost control solenoid duty goes from 5% to 95% in one big step, and the pressure goes up. Not very useful to detect a problem in the mid range of actuator travel.

The swirl valve also tests out ok, and i can rule this out because it is closed at full throttle (e.g. in the 6th gear recording it was long closed before the spike and drop).

If someone can read HDS recording files, i can provide the originals. Perhaps someone knows a good honda garage which could look at the files an provide me with some hints - the local garage does not know much about the diesel, and based on the big screw up with my clutch (got oil on CG Kevlar clutch, as confirmed by CG), i do not trust them anymore.

I will play this step by step with an independant mechanic, i have access to an HDS myself to carry out tests and recordings.
 

Attachments

·
Cockup Specialist
Joined
·
12,885 Posts
Or it could be the EGR valve dragging its heels and venting more and more boost as it opens slightly releasing the exhaust pressure on the turbine ?
Hence VNT position the same (solenoid duty cycle) but boost pressure drops.
Basically is the EGR staying open and drifting.
Have you cleaned and Flushed the EGR and swirl valves ?

Still looks like a boost leak venting access pressure....but if there is no leak in pipes, gaskets or intercooler then it must be the EGR ?
Gets some soapy water on all the joint pipes and intercooler and look for bubbles.
Preferably on a hub dyno where you can put it under load.
 

·
Premium Member
Joined
·
131 Posts
Discussion Starter · #29 ·
According to my service manual, the EGR valve is designed in a way that as long as intake pressure (boost) is higher than the exhaust pressure before the turbo, it will stay closed by force of intake pressure, even if the spring was broken and the valve would move freely.

The only condition where i think it could be possible to force the valve open is if exhaust pressure gets high enough to overcome the boost pressure in the intake manifold (and overcome the spring which keeps the valve shut unless vaccuum is applied to the diaphragm, provided it is not broken).

Then the turbo would loose some exhaust volume, and boost would drop. As far, i could not yet find a definite pressure value where this would happen, sometimes i get overboost, sometimes it starts with a pressure drop also (mos common in 5th/6th gear when i am at light throttle and start pressing the pedal down).

Also, with the EGR opening uncommanded, i should get a significant drop in volumetric efficiency (e.g. from 80% down to 40%, like you logged some time ago), because some air volume is replaced with exhaust gas, right? But this is not happening, VE stays at full throttle values.

Perhaps the dreaded exhaust manifold crack is a more likely explanation? This would also cause a loss of exhaust volume for the turbo, but it would not be registered by the MAF sensor, because the engine got the full air volume, but some of it is not available to power the turbo.
 

·
Cockup Specialist
Joined
·
12,885 Posts
Yeah your right about the EGR to some degree unless its not closing/seating fully due to debris or coke.
But VE would be hit like you say.

Check for leaks everywhere then.
Did you also check the airbox to turbo duct at the back....in case the MAF reading is innaccurate.
 

·
Premium Member
Joined
·
131 Posts
Discussion Starter · #31 ·
I have now checked the DPF differential pressure with the HDS - it is at 31 kPa (310 mbar) at 4030 rpm at full throttle.

This is with 5.4 g soot in the filter according to the HDS, EGT 1 (before warm up catalyst) 476°C, EGT 2 (before DPF) 392°C, MAP at 2.06 bar, MAF 139,7 g/s, IAT 1: 20°C (measured at MAF sensor), IAT 2: 49°C (measured after intercooler by MAP sensor), injection quantity 56.5 mg/stroke, MAF per stroke is 1040 mg, this means AFR is 18.4.

I was lower (22 kPa) in Winter, but with a nearly empty DPF (0.7 g soot). At least i get the same injection quantity as in my winter log.

Is there too much backpressure?
 

·
Premium Member
Joined
·
131 Posts
Discussion Starter · #34 ·
1040mg at 4000rpm is normal, your IAT2 is good. Have you got FMIC??
I have HDS and the stock AFR at 4000rpm is 19,5:1 (53-53.5mg IQ).
No, i have just the standard intercooler. It was about 18°C outside in the evening, after about a minute of 50 km/h driving through a small village. A/C was off. IAT 2 will go up to 60°C when it is warmer or during longer spirited driving.

I also thought the AFR would be limited to 19,5:1 at 4000 rpm, after looking at an OEM map download i found on the net with WinOLS. However, i have a very old software version, which might be different.

I think this recording is just on the edge where the smoke limiter would kick in, a few degrees warmer or some 0.05 bar less boost and IQ would have been limited.
 

·
Registered
Joined
·
7 Posts
Where does one buy the HDS ?
Here for example:
Honda Diagnostic System HDS HIM
There are other similar sites.

No, i have just the standard intercooler. It was about 18°C outside in the evening, after about a minute of 50 km/h driving through a small village. A/C was off. IAT 2 will go up to 60°C when it is warmer or during longer spirited driving.

I also thought the AFR would be limited to 19,5:1 at 4000 rpm, after looking at an OEM map download i found on the net with WinOLS. However, i have a very old software version, which might be different.

I think this recording is just on the edge where the smoke limiter would kick in, a few degrees warmer or some 0.05 bar less boost and IQ would have been limited.
Summer in Spain IAT +50-55ºC over the temperature outside:D, with FMIC +22º.

And yes, in all ecus I have seen with winols driver wish is 53mg IQ and lambda map at 4000rpm 19,5:1.
 

·
Premium Member
Joined
·
131 Posts
Discussion Starter · #36 ·
I got mine from a german Honda forum user. Was apparantly also bought somewhere online, and looks exactly like the one you posted. He sold it together with a Dell D600 notebook with 1,6 Ghz Pentium M an 1,5 GB of RAM. The HIM needs a COM port for communication with the computer, although a USB to serial converter might work too, although i did not test it.

I see an FMIC would limit my IATs to about 50°C even in hottest summer weather (does not get above 30°C here), and to 40°C in normal weather. That would mean the smoke limiter would not be triggered in any normal conditions, even with boost below the target value.

I think now we have the reason why there are performance gains with an FMIC even with a stock ECU (where drivers wish/torque limiter should be setting the IQ regardless of IAT as long as there is enough air mass and no IAT limits are exceeded) - the original intercooler is too small to avoid running into the smoke limiter when the IAT is too high. Also, as far as i have seen, the altitude dependant boost limiter cuts in below 950 mbar ambient pressure (roughly 550 meters at standard pressure), which will limit airflow too.

Maybee i will have to look for an FMIC, once the boost control issue is resolved. This would certainly also help the DPF (less soot produced due to leaner AFR).
 

·
Premium Member
Joined
·
131 Posts
Discussion Starter · #38 ·
That's what i think too. I hope it is just the solenoid which controls the vaccuum to the actuator, and not the actuator/diaphragm itself. Because then i guess it would mean i need a new turbo, as you do not get the diaphragm or actuator as a separate part (at least not officially).

A friend had a similar problem with a Seat some time ago, there it was just the soleniod. I have arranged a checkup with a knowledgleable mechanic in about a week, then i will hopefully know more.
 

·
Premium Member
Joined
·
131 Posts
Discussion Starter · #39 ·
At the moment, it looks like the EGR is forced open at high boost. We tried blocking the EGR with a blanking plate, and the jerk in acceleration when i floor it aggressively was gone - at least, i could not get it to appear on 10+ trials, at 750 meters altitude and 20°C - conditions where i would normally always get it.

I have, or at least thought i have, an EGR delete (was done when the software was downgraded), but unfortunately it is not complete. Although the duty cycle of the valve is low enough to keep it closed under normal conditions, it still opens at idle, and its function is still checked by the ECU.

I could get away with the blanking plate as long as i kept driving, and did not let the engine idle for more then 12 seconds with the car stopped. Also, we found out that the "old software" appears to be in fact a remap, with the EGR not deleted correctly.

Now, we will start from scratch with an original file, try to get rid of the EGR here (another tuner), and see if it works together with the blanking plate (no faults, pressure problems gone), then restore the parameters of the tuned file to get the power back where it was.

Can anyone guess how much BHP i could have had with 56.5 mg injection quantity at 4000 rpm and an advanced injection timing of about 2°? A road dyno program reads about 160 BHP under ideal conditions (with generic assumptions on drivetrain losses, air resistance and the offical quoted weight of the car), calculations based on airflow and IQ alone (disregarding the timing) come between 155 to 160 BHP DIN corrected.

I do not want to go too far because of the clutch and the DPF, so it think around 160 - 165 BHP and 370 Nm should be rather safe. In fact, with the actual remap, the DPF regeneration intervalls where nearly twice as far apart as before (550 - 860 km, vs. 330 - 450 km), but it think this was an effect of the partial EGR delete.

If we cannot get it working, i think this will be a job for Fahad, but i will have to talk about this with the mechanic who does the flashing if he accepts (or can use) such a file.
 

·
Cockup Specialist
Joined
·
12,885 Posts
1040 mg/stroke MAF
53.25 mg/stroke MFF
19.53 AFR @ 4000rpm
1040 * 2 * 4000 = 8320000 mg/min
8320000 / 453592.37 = 18.342 lb/min
0.38/60 = BSFC

(lb/min) / (BSFC/60 * AFR) = BHP
18.342 / (0.38/60 * 19.53) = 148.29 bhp @ 4000rpm (150.35ps)
 
21 - 40 of 50 Posts
Top