[mvadg]

Originally Posted by arun francis

Hi, My name is Arun and I am driving a 2010 corolla.
Nowadays all cars are having fuel efficiency indicators, ie they show average fuel consumption, instantaneous fuel consumption, distance to empty etc.
But I wonder whether we can really rely on it.
In order to show the correct fuel consumption the ecu or on board computer has to monitor the actual fuel consumption by the engine, this can be done by fixing flowmetres in fuel lines.
But I feel that most manufacturers are putting a software/algorithm in the ecu, so that the fuel consumption is calculated by taking into account engine speed and vehicle speed.
I invite other BHPians to share their knowledge in this regard.

The ECU also controls the amount of fuel injected and distance traveled, so it does really know the actual fuel efficiency.

If you read up a bit on the data collected by the various parts of the CU - like the ECU, BCU, TCU you'll be astounded!

GIYF (and may be evil too ).

[Jeroen]

Originally Posted by mvadg

The ECU also controls the amount of fuel injected and distance traveled, so it does really know the actual fuel efficiency.

If you read up a bit on the data collected by the various parts of the CU - like the ECU, BCU, TCU you'll be astounded!

GIYF (and may be evil too ).

Agree, the ECU knows pretty well and pretty accurate on how much fuel it injects. You would have to use at least two very accurate and thus expensive, flowmeters to get a more accurate measurement I would think. One in the supply to the engine and one in the return line for most engines.

Jeroen

[tejas08]

With advancement in Modern Day MIDs and lot of other Electronics Stuff, we are getting much better details in terms of Fuel Economy and stuff. Multiple Trip Meters allow users to track different trips which helps people in comparing MIDs Vs. Tankful to Tankful Method.

The question now arises as to what is the current accuracy levels of the MIDs Vs. Tankful to Tankful Method. Here's one of the link's where a lot of discussions on this topic has happened.

http://www.team-bhp.com/forum/techni...lay-mid-9.html

To start of the discussion, in my 4th Gen City I am getting a difference of less than 3% between MIDs and Tankful to Tankful Method or should I say Tankful to Tankful method gives me 97% Fuel Economy of that of the MID. So if I have a FE of 10 kmpl on the MID, the amount of fuel that goes in would lead me to a FE of somewhere between 9.7 to 9.8 kmpl.

[4x4power]

From my personal experience the FE figures from MID are not quite exact. There will be a significant difference between the displayed FE and actual FE. Honestly, I won't trust MIDs in case FE. But for other informations MIDs are very helpful. As written above, the loads of information depends upon the price of the car. So there will a lot of useful infos displayed on higher priced cars. But unfortunately I haven't got the chance to drive one.

[Surya-TJet]

A lot of people already know how the fuel efficiency display is calculated, but as a small precursor to the uninformed:

Fuel economy display in the MID is based on the input from the ECU. ECU is the one which controls the amount of fuel being squirted into the intake manifold (incase of mpfi petrol engines) or the cylinder head (incase of crdi diesels, unit pump diesels or direct injection petrol engines). Now, for each power stroke of the engine, the ECU decides on the quantity of fuel to be sprayed by each injector based on various factors listed below:

1. Acc. Pedal position
2. Engine Load
3. gear position
4. intake manifold pressure
5. exhaust gas temperature
6. inlet air temperature
7. turbocharger boost pressure
8. inputs from traction control module
9. Inputs from Transmission control module (incase of AMT/AT/LSD/4x4)

The gearbox output shaft in the car has a sensor which gives inputs based on no. of pulses / km (either light based electrical pulses or worm-wheel electromagnet based).

The ECU starts keeping track of the no. of pulses once the vehicle starts moving. It simultaneously starts keeping track of the fuel injected in each cylinder during each power stroke. This will help in calculating total fuel injected (in terms of milligrams / stroke) for the distance covered.

Then after covering 1 km (as a general rule of thumb), it starts to display the figure on the MID. This figure will always use a "correction factor" which is pushed entirely by the marketing department.

For example, say the car about to be launched has got a slightly lesser fuel economy than the segment leader, then the "correction factor" is increased from say 1.02 to 1.07 or 1.08. This correction factor is multiplied to the value calculated by the ECU and then displayed in MID.

A few modern cars have an "instantaneous readout" in the MID in terms of L/100 kms or km/l. This is again given with the "correction factor". Generally, the MID is just a marketing gimmick, nothing more. The fuel economy should remain consistent within a given range between tankful to tankful driving. Only if one observes a sudden drop in fuel economy compared to their previous drive, should we be worried about the engine and fuel system.

For those of you wanting to extract the best fuel efficiency while driving can do so by adopting "defensive driving techniques" instead of relying on some digital output from a "Marketing Driven" MID.

P.S - I am not against technology or any manufacturer in general, just that it should be really useful technology like ABS/ Airbags / EBD / TCS / ASR instead of some silly fuel economy readout forced by marketing blokes.

Just my two cents.

Cheers & Safe Driving.

[Midas]

Originally Posted by Surya-TJet

the distance covered.

Then after covering 1 km (as a general rule of thumb), it starts to display the figure on the MID. This figure will always use a "correction factor" which is pushed entirely by the marketing department.

Any reliable source for this 'correction factor'? If they are really using some multiplication factor like this, then it's really unethical and can be termed as cheating. I guess it's even legally possible to sue manufacturers for such lies.

[mxx]

Originally Posted by Surya-TJet

This figure will always use a "correction factor" which is pushed entirely by the marketing department.
.

What is the source for this information?

[Surya-TJet]

Originally Posted by Midas

Any reliable source for this 'correction factor'? If they are really using some multiplication factor like this, then it's really unethical and can be termed as cheating. I guess it's even legally possible to sue manufacturers for such lies.

Originally Posted by mxx

What is the source for this information?

I am telling this from multiple user experiences when using OBD diagnostic tools (not your average ELM 327 adapters, but the highly accurate ones which are manufacturer specific).

We calculate data based on fuel consumption quantity in terms of "mg/stroke" for both petrol & diesel cars. This is the data which is then used by the ECU to calculate the short term fuel trim, long term fuel trim and provide data on the fuel consumption.

I have used such tools across various platform of vehicles (cars, trucks, buses, motorcycles, defence vehicles).

I unfortunately cannot share any further details. But, as I said earlier, it is always shown as a marketing gimmick.

In certain CAN system based cars which can digitally monitor and track the actual fuel quantity inside the tank at that instant, this fuel flow calculation algorithm is also used to detect and warn fuel leakage so that any fire accidents / hazards can be avoided.

Based on the car industry and more recently the truck industry, I have observed that more and more manufacturers are introducing a "correction factor" to the displayed value, primarily due to pressure from the marketing department.

Also, your vehicle speed sensor and odometer sensor have an allowed positive error margin up to 10% in India. Also, if the customer changes his tire size, it again changes the accuracy of these values as these sensors are calibrated for one particular tire size (hence the 2% prescribed error limit during tire swap; although that is also to avoid suspension geometry change and avoid poor braking performance).

This is why there is a difference in the fuel economy calculation between tank-to-tank method and the values displayed by the MID.

Hope this helps..!! Cheers & Safe Driving..!!

[StarrySky]

I can totally believe that such correction factors exist. In my previous car (Punto in India), the correction factor was probably quite low. My tankful-to-tankful mileage used to be very close to the MID value - maybe about 98% accuracy overall, to put a number on it to answer the poll question.

In my current car (Octavia), the accuracy is about 90%, falling even further in winter months (worst was about 75% accuracy). Not a surprise, given the recent controversies surrounding VAG

[Jeroen]

Originally Posted by Surya-TJet

I am telling this from multiple user experiences when using OBD diagnostic tools (not your average ELM 327 adapters, but the highly accurate ones which are manufacturer specific).

Very interesting. I have a several OBD diagnostic tools, from generic handscanners to PC based one. Could you give me an example where you get a more accurate reading based on an “accurate manufacturer” specific analyser compared to a more generic one. I get more data with the manufacturer specific one, not necessarily better or more accurate data. I would appreciate if you could explain what data you get and why it is more accurate with your manufacturer specific tool

Originally Posted by Surya-TJet

We calculate data based on fuel consumption quantity in terms of "mg/stroke" for both petrol & diesel cars. This is the data which is then used by the ECU to calculate the short term fuel trim, long term fuel trim and provide data on the fuel consumption.

short term and or long term fuel trim have nothing to do with the fuel consumption perse. At least not in measuring it. Principle the fuel quantity gets calculated by the ECU as it know the fuel pump delivery and injector characteristic and the injection period. See Der Alte’s explanation in some more detail earlier on in this thread. Fuel trim is calculated by the ECU in order to inject the correct amount of fuel. On modern car there are quite a few other parameters that get measured and are part of the ECU calculation to determine how much fuel needs injecting. But the amount of fuel (and therefor fuel usage) is (physically) determined by pump, injector and injection period.

Originally Posted by Surya-TJet

I unfortunately cannot share any further details. But, as I said earlier, it is always shown as a marketing gimmick.

Why not, what so secret?

Originally Posted by Surya-TJet

In certain CAN system based cars which can digitally monitor and track the actual fuel quantity inside the tank at that instant, this fuel flow calculation algorithm is also used to detect and warn fuel leakage so that any fire accidents / hazards can be avoided.

That would be an interesting trick. The biggest problem is see is that the actual sensors used to measure the fuel quantity inside the tank are pretty pathetic. No matter how much digitally monitoring and CANs you throw at it, it remains a crappy sensor system. It would also have to be phenomenally accurate and fast to provide a meaningful fuel leakage warning. Not an easy task in a moving tank with fuel sloshing around the tank and not to accurate a sensor to start with.

I’m pretty familiar with how the aviation industry detects fuel leakage on planes, but I have yet to see it being done on cars. Could you share some more information on these systems and which manufacturers install such systems.

Originally Posted by Surya-TJet

Based on the car industry and more recently the truck industry, I have observed that more and more manufacturers are introducing a "correction factor" to the displayed value, primarily due to pressure from the marketing department.

My own observation is that MID tends to be on the cautious side. So they will tell you your tank is empty before it really is. If anything it’s probably more the legal department then the marketing department that is involved here. Once you have bought the car and it doesn’t do what has been advertised/spec-ed it (can) become a legal matter. And in some countries it does become legal. One of the main reason speedometers are overly optimistic. That way, if you keep to the speed limit, you are always a bit slower.

Originally Posted by Surya-TJet

Also, your vehicle speed sensor and odometer sensor have an allowed positive error margin up to 10% in India. Also, if the customer changes his tire size, it again changes the accuracy of these values as these sensors are calibrated for one particular tire size (hence the 2% prescribed error limit during tire swap; although that is also to avoid suspension geometry change and avoid poor braking performance).

Just out of curiosity, is the 10% error margin a legal requirement. So do they get tested/verified against that margin?

Originally Posted by Surya-TJet

This is why there is a difference in the fuel economy calculation between tank-to-tank method and the values displayed by the MID.
!

Yes, there is. Mainly based on, I would think, legal considerations. And hopefully a few common sense people at the automotive industry who think in general it is prudent to be on the safe side of cautious.

The MID might be a marketing gadget as such of course. I like them, as i like fiddling with buttons, menu’s etc.

Jeroen

[Surya-TJet]

Originally Posted by Jeroen

Very interesting. I have a several OBD diagnostic tools, from generic handscanners to PC based one. Could you give me an example where you get a more accurate reading based on an “accurate manufacturer” specific analyser compared to a more generic one. I get more data with the manufacturer specific one, not necessarily better or more accurate data. I would appreciate if you could explain what data you get and why it is more accurate with your manufacturer specific tool

In manufacturer specific scanners (2 out of the many I have used), one can change the refresh rate of data acquisition on a recording device (like a data logger) and then review it later based on individual sequences / timestamps (hence finer sampling). This will help in understanding the changes just before or after a certain abnormality was noticed in the engine / drivetrain and possibly zero-in on the causative factor for the said failure.

Originally Posted by Jeroen

short term and or long term fuel trim have nothing to do with the fuel consumption perse. At least not in measuring it. Principle the fuel quantity gets calculated by the ECU as it know the fuel pump delivery and injector characteristic and the injection period. See Der Alte’s explanation in some more detail earlier on in this thread. Fuel trim is calculated by the ECU in order to inject the correct amount of fuel. On modern car there are quite a few other parameters that get measured and are part of the ECU calculation to determine how much fuel needs injecting. But the amount of fuel (and therefor fuel usage) is (physically) determined by pump, injector and injection period.

It is not directly involved in measurement, but in tracing the potential fuel leaks over a preset time duration "IF" the manufacturer has programmed the same into the system.

See, an engine is designed for consuming only a certain maximum quantity of fuel for a given power rating. That is why the related engine components like fuel injectors and fuel filters have a maximum flow rate that they can effectively handle.

Once the short term / long term values show inconsistent AFR values in comparison with the maximum / minimum recommended limits in the ECU programming to ensure emission adherence of the engine and if the ECU is programmed to note this abnormality, it can be used to highlight leaks in the fuel system (some parts in the fuel system which have potential for fuel leaks are fuel filter, injector high pressure lines, rail pressure sensor end connections, common rail-injector line flange etc).

I had the unique opportunity of facing this situation in real life when a vehicle that I was testing just did not generate sufficient power even when the driver depressed the accelerator pedal fully (maximum torque demand). There was no real abnormality in all the sensor readouts, except for the Short term and long term fuel trims, which were off the charts. We then did a detailed examination in the engine bay and found that the high pressure inlet line from the common rail to the injector body had a micro-crack due to manufacturing defect (found later from metallurgical study), which caused significant fuel leak at that spot when the ECU tried feeding more fuel by raising the rail pressure significantly compared to normal engine operating conditions.

Although the system as such would not have the capability of highlighting the point of leak, it will atleast be able to give an error message to the driver on the MID, which might help him from potentially getting stranded with a broken down car in the middle of nowhere at night.

This is again purely manufacturer dependent to provide such systems and not a mandate (at least not here in India, no idea about other places).

Originally Posted by Jeroen

Why not, what so secret?

I cannot discuss about this in a public forum in any further detail, as it pertains to my job and its security. I tried to send you a PM on this one, but since you prefer not receive any, lets leave it here.

Originally Posted by Jeroen

That would be an interesting trick. The biggest problem is see is that the actual sensors used to measure the fuel quantity inside the tank are pretty pathetic. No matter how much digitally monitoring and CANs you throw at it, it remains a crappy sensor system. It would also have to be phenomenally accurate and fast to provide a meaningful fuel leakage warning. Not an easy task in a moving tank with fuel sloshing around the tank and not to accurate a sensor to start with.

I’m pretty familiar with how the aviation industry detects fuel leakage on planes, but I have yet to see it being done on cars. Could you share some more information on these systems and which manufacturers install such systems.

I did not mention anywhere that such measurement systems are in place in all production cars. The quality and resolution of the fuel sensor is of a separate debate relating to the cost targets to be met against each system in a vehicle.

I have seen the use of highly accurate fuel system sensors in race cars and rally cars, not on road cars (even the uber-luxury ones like MayBach / Rolls Royce).

The tank slosh is taken care of by a pressurized system (again race cars), where the slosh is limited to a great extent to ensure that every last of drop of fuel can be used effectively without any air suction into the fuel lines under extreme "G"-Forces. I have not seen such systems in real life, but only have read / referenced some related information from some "closed sources".

I genuinely do not know how it is in aviation industry, but if you feel you can share come basic / generic information, I'd be thankful for having learnt some more useful information in my life.

Originally Posted by Jeroen

My own observation is that MID tends to be on the cautious side. So they will tell you your tank is empty before it really is. If anything it’s probably more the legal department then the marketing department that is involved here. Once you have bought the car and it doesn’t do what has been advertised/spec-ed it (can) become a legal matter. And in some countries it does become legal. One of the main reason speedometers are overly optimistic. That way, if you keep to the speed limit, you are always a bit slower.

That part I do realize, but it will in no way be relevant to the instantaneous and average fuel economy readout on the dashboard. I still remember the Top Gear show where Jeremy Clarkson drove a Diesel Jaguar from Switzerland to Blackpool with the car's range indicator showing "0 Miles" and upon inspection at the finish line, they still found quite a bit of fuel left inside the tank.

I really do not know how the legal system is in other countries, but here in India, the laws are always in favour of the corporate giants and not in support of the common man. Hence, you can see some Car Manufacturer's CEO has the audacity to openly admit to all media that their cars need not be necessarily safe to meet the crash norms of "XYZ" and that it is sufficient for them to meet the regulatory guidelines as laid down by the government.

Originally Posted by Jeroen

Just out of curiosity, is the 10% error margin a legal requirement. So do they get tested/verified against that margin?

Yes, there are norms as laid out by the CMVR. And yes, every manufacturer has to prove that their speedometer / odometer is calibrated in such a manner that they are between 0.1% - 10% positive error margin. Only then they are certified to sell that particular model of vehicle. They also have to prove this upon random sampling for Conformance to Production (COP).

Originally Posted by Jeroen

Yes, there is. Mainly based on, I would think, legal considerations. And hopefully a few common sense people at the automotive industry who think in general it is prudent to be on the safe side of cautious.

Trust me, there are next to no laws in India which will point an automotive manufacturer as "GUILTY". Multiple examples to prove this same point (GM Tavera Emission Scandal, VW Diesel Gate, Skoda's ASS Horror Stories and their litigative torture to a customer (BHP-ian), Toyota's airbag malfunction etc). Lots of such threads exist in Team-BHP only to reiterate the same.

Originally Posted by Jeroen

The MID might be a marketing gadget as such of course. I like them, as i like fiddling with buttons, menu’s etc.

Jeroen

The MID I feel should serve as a basic assistant during driving and not become the core-detractor for the driver. Unfortunately here in India, we see the exact same happening on the highways (cars driven at 50kmph on high speed lanes in 6 lane highway under the name of fuel economy, etc.)

In this regard, I like how certain manufacturers don't allow drivers to fiddle too much with the MID when the vehicle is in motion (Hyundai is a prime example, although their MID never shows fuel economy). This helps in maintaining driver focus on the road and therefore enabling safer driving for themselves as well as to others on the road.

Cheers & Safe Driving..!!

[Jeroen]

Originally Posted by Surya-TJet

In manufacturer specific scanners (2 out of the many I have used), one can change the refresh rate of data acquisition on a recording device (like a data logger) and then review it later based on individual sequences / timestamps (hence finer sampling). This will help in understanding the changes just before or after a certain abnormality was noticed in the engine / drivetrain and possibly zero-in on the causative factor for the said failure.

As I said earlier, you tend to get more data, not more accurate data perse.

Originally Posted by Surya-TJet

Once the short term / long term values show inconsistent AFR values in comparison with the maximum / minimum recommended limits in the ECU programming to ensure emission adherence of the engine and if the ECU is programmed to note this abnormality, it can be used to highlight leaks in the fuel system (some parts in the fuel system which have potential for fuel leaks are fuel filter, injector high pressure lines, rail pressure sensor end connections, common rail-injector line flange etc).

If the fuel trim goes whacky there could be a whole host of reasons, a leak being one of the least likely. Also, there is no way I can think of, maybe you can, how you could distinguish between whacky fuel trims for all sort of reasons versus whacky trims due to a fuel leak.

Originally Posted by Surya-TJet

I had the unique opportunity of facing this situation in real life when a vehicle that I was testing just did not generate sufficient power even when the driver depressed the accelerator pedal fully (maximum torque demand). There was no real abnormality in all the sensor readouts, except for the Short term and long term fuel trims, which were off the charts. We then did a detailed examination in the engine bay and found that the high pressure inlet line from the common rail to the injector body had a micro-crack due to manufacturing defect (found later from metallurgical study), which caused significant fuel leak at that spot when the ECU tried feeding more fuel by raising the rail pressure significantly compared to normal engine operating conditions.


Although the system as such would not have the capability of highlighting the point of leak, it will atleast be able to give an error message to the driver on the MID, which might help him from potentially getting stranded with a broken down car in the middle of nowhere at night.

See my point before. I doubt very much that any such system could provide a warning you might have a leak. In case of severe leaks you might see a Check Engine light or similar popping up, but that is most likely due to the fact that your air fuel ratio is of the charts, not because of the leak.

Correlating a fuel leak to fuel trim(i.e. a set of parameters and or conditions) seems not realistic to me at all.

Originally Posted by Surya-TJet

That part I do realize, but it will in no way be relevant to the instantaneous and average fuel economy readout on the dashboard. I still remember the Top Gear show where Jeremy Clarkson drove a Diesel Jaguar from Switzerland to Blackpool with the car's range indicator showing "0 Miles" and upon inspection at the finish line, they still found quite a bit of fuel left inside the tank.

I remember that, it was hilarious. My Jaguar isn’t quite that bad, but when my range indicator hits zero I can still drive well over 100km. And I do, drives my poor wife bonkers.

Originally Posted by Surya-TJet

I really do not know how the legal system is in other countries, but here in India, the laws are always in favour of the corporate giants and not in support of the common man. Hence, you can see some Car Manufacturer's CEO has the audacity to openly admit to all media that their cars need not be necessarily safe to meet the crash norms of "XYZ" and that it is sufficient for them to meet the regulatory guidelines as laid down by the government.

Thanks. In many western countries there tends to be (strong) consumer protectiveness build into the legal system

Originally Posted by Surya-TJet

Yes, there are norms as laid out by the CMVR. And yes, every manufacturer has to prove that their speedometer / odometer is calibrated in such a manner that they are between 0.1% - 10% positive error margin. Only then they are certified to sell that particular model of vehicle. They also have to prove this upon random sampling for Conformance to Production (COP).

Thanks.

Jeroen

[Seacat]

Hi ,
Have noticed since past 3 fuel refills that the Tucson instantaneous fuel consp on the dash shows 8.4 km/ltr whereas the ave fuel cons upon completing the trip shows 10.4 km/ltr. I find a disparity of about 2 km/ltr or more at times between these two figures.
I have been also noticing that normally when the tank is full the instant fuel consp on the dash shows a much higher figure (sometimes even 11km/ltr) in city but then when the tank is say half full it shows a consistent figure irrespective if the road is empty and traffic very light.

Is it possible to get the instantaneous fuel consumption calibrated/corrected to reflect accurately and how ?

Regards,