Go Back   Team-BHP > Under the Hood > Technical Stuff


View Poll Results: What percentage accuracy do you get comparing MID to Tankful to Tankful Method
99% & above 0 0%
97% to 99% 4 16.67%
94% to 97% 2 8.33%
90% to 94% 7 29.17%
Below 90% 4 16.67%
Never Checked 7 29.17%
Voters: 24. You may not vote on this poll

Reply
 
Thread Tools Search this Thread
Old 11th December 2010, 22:58   #16
BHPian
 
Dr.AD's Avatar
 
Join Date: Apr 2010
Location: Bangalore/Pune
Posts: 428
Thanked: 1,091 Times
Default Re: How accurate is Multi Information Display (MID)

I could be completely wrong on this one, but this is my hypothesis based on my experiences with cars with MID and manual "tank full to tank full" measurements:

If your idling time during a particular trip is negligible (e.g. on a highway drive) then the MID numbers are very close to the manual numbers. Somehow, and in some cars, the MIDs get confused during idling. The instantaneous fuel consumption during idling is in reality 0 km/l, which the MID shows either as "---" or "N/A". Now, the interesting point is that how is this used in average mileage calculations? It should be used as 0 km/l in average calculations. If this is not used or simply ignored, then the average numbers will start going astray. My observation is that a big amount of idle times means a bigger difference between the MID numbers and the manual numbers. Maybe some of the MID algorithms simply ignore the idle times (when the display shows "----"), which leads to more optimistic average numbers.

Again, this is just a hypothesis, and I do not have any data to prove this (and hence I call this a hypothesis).
Dr.AD is offline   (1) Thanks Reply With Quote
Old 12th December 2010, 21:15   #17
Senior - BHPian
 
DerAlte's Avatar
 
Join Date: Mar 2007
Location: Bangalore
Posts: 8,076
Thanked: 2,876 Times
Default Re: How accurate is Multi Information Display (MID)

@trust-in-thrust, more sophisticated need not mean more expensive. It is only that design time penny-pinching saves a lot for ECU manufacturers. Sale price of the ECU (ECU manu -> Car manu) depends on the (software) features delivered with the ECU. Hardware price is relatively constant. Software feature price = opportunity cost, since development effort would have paid itself back.

So, if there is no MID, that feature is not delivered. Not delivered = not present in the ROM, and not 'present in the ROM but disabled'.

There are more accurate instrumentation available - being in Automation, you should know . An accurate flow-meter justifies it's cost if the clients can see the benefits of consumption trends. I am not assuming ECU control in the engines that you were visualizing for this.

@Dr.AD sir, consumption at idling is not 0 km/l, nor an insignificant figure. What really happens is that the figure is small enough that one needs to accumulate it for a long time to get a significant figure in the MID. Your premise is right, that highway figures are more accurate than city start-stop driving figures with a lot of idling.

Last edited by DerAlte : 12th December 2010 at 21:20.
DerAlte is offline   (1) Thanks Reply With Quote
Old 12th December 2010, 21:56   #18
BHPian
 
Dr.AD's Avatar
 
Join Date: Apr 2010
Location: Bangalore/Pune
Posts: 428
Thanked: 1,091 Times
Default Re: How accurate is Multi Information Display (MID)

Quote:
Originally Posted by DerAlte View Post
@Dr.AD sir, consumption at idling is not 0 km/l, nor an insignificant figure.
I am not sure what you mean by consumption at idling not being 0km/l. When you are idling, you are consuming fuel but not moving, and thus, theoretically, the instantaneous consumption is indeed 0km/l.

Quote:
What really happens is that the figure is small enough that one needs to accumulate it for a long time to get a significant figure in the MID.
This is what I was trying to say. The idling consumption of 0km/l for that time must be accumulated (and not simply ignored by the excuse of sampling data N/A) to get an overall accurate average consumption. I was just wondering if all cars do this or if they simply ignore the idling data. This would corroborate the observation that the MID figures differ from true average consumption when there is a lot of idling involved.
Dr.AD is offline   Reply With Quote
Old 12th December 2010, 23:26   #19
Senior - BHPian
 
DerAlte's Avatar
 
Join Date: Mar 2007
Location: Bangalore
Posts: 8,076
Thanked: 2,876 Times
Default Re: How accurate is Multi Information Display (MID)

Dr.AD, your literal '0 Km/l' at idling (with the vehicle not having moved an inch) is, quite literally, ignoring the fact that 'consumption' makes sense only over a large period of time, during which the vehicle would have stopped, idled, moved, stopped, idled ... If one intra-polates the same for every stroke of the engine or every Km traveled, one would get a different picture and lose sight of the woods for the trees.

No, ECUs that calculate the 'average' accumulate the injection data - no matter whether the vehicle is moving or not. This also accounts for periods when the vehicle is 'coasting', i.e. not using fuel at all, e.g. going downhill.

IMHO, the issue is that of calculating with a stable, known, irrefutable reference. 'Full tank to full tank' has its inherent errors. But then, almost no one says 'I get 11.67 Kmpl', instead they approximate it to 'gyarah-barah deti hai'!
DerAlte is offline   Reply With Quote
Old 14th December 2010, 19:34   #20
BANNED
 
Join Date: Nov 2008
Location: New Delhi
Posts: 690
Thanked: 152 Times
Infractions: 0/1 (5)
Default Re: How accurate is Multi Information Display (MID)

@deralte: It is not true that hardware cost will remain the same.We need to see the processing capability of the ECU .It is also not true that ECU will cost the same and you just need to enable few software features. At the same time yes software cost is also a major component.

What i mean to say is the more the data you have the greater the cycle time(read higher time for executing each instruction bit) with the same hardware .
This is the only reason why you do not find advanced MID's in many cars like the city for eg as it will mean upgrading the board components of ECU making the ECU expensive (add cost of mid interface , wiring) .

Also note if you try to push more software on the same ECU you will end with slow response from ECU for inputs which mean a lot like throttle response, fuel rail pressure , injection cycles etc.


Regards

Last edited by Trust_In_Thrust : 14th December 2010 at 19:37.
Trust_In_Thrust is offline   Reply With Quote
Old 14th December 2010, 20:13   #21
Distinguished - BHPian
 
Join Date: Jul 2009
Location: Kolkata
Posts: 3,506
Thanked: 2,068 Times
Default Re: How accurate is Multi Information Display (MID)

Quote:
Originally Posted by DerAlte View Post
....
IMHO, the issue is that of calculating with a stable, known, irrefutable reference. 'Full tank to full tank' has its inherent errors. But then, almost no one says 'I get 11.67 Kmpl', instead they approximate it to 'gyarah-barah deti hai'!
Hi,
So (assuming honest dealer) between the MI Display and our rough and ready tank full to tank full method, we should always go by the MID?

Regards
Sutripta
Sutripta is offline   Reply With Quote
Old 2nd September 2011, 22:01   #22
BHPian
 
Tassem's Avatar
 
Join Date: Sep 2009
Location: Bangalore
Posts: 467
Thanked: 425 Times
Default Re: How accurate is Multi Information Display (MID)

Sorry to revive an old thread but I didn't want to create a new one.

I recently noticed two things about the Average Consumption MID on the Verna.

1- Highway Accuracy is much higher compared to City Accuracy.

2- If I top up to a full tank from half tank or more, the accuracy is higher, compared to topping up when the fuel is less than half tank.

When I say accuracy, the city reading has been off by 2.5-3kmpl, while the highway accuracy has always been within .5kmpl.

I can understand the first bit, perhaps the idling and coasting confuses the MID. But what's with the half tank phenomenon?

Has anyone else experienced this?

I tried about 6 top ups and this is not a flash in the pan. It's the same everytime. I also stick to the same nozzle and tell the guy not to top up after the auto shut off kicks in.

Any views?
Tassem is offline   Reply With Quote
Old 3rd September 2011, 01:25   #23
BHPian
 
Join Date: Aug 2011
Location: Nasik / Mumbai
Posts: 427
Thanked: 591 Times
Default Re: How accurate is Multi Information Display (MID)

OK.. the probable reason for the difference in the MID shown value and the Actual value in the city is probably due to the evaporation in course of usage of the fuel. When being used in the city, the same 'tankful of fuel' stays in the car for a lot more days as compared to when you drive on the highway (i.e. you finish the 'tankful of fuel' faster when driving on the highway in terms of days taken). When the same tank of fuel stays in the tank for more days, obviously, it will get more time to get evaporated when you park the car in the sun everyday.

This loss of fuel to evaporation is not counted by the MID (which only calculates the fuel entering the combustion chamber) and hence there is a greater difference when we compare FE differences in the city as compared to those on the highway.

And regarding the half tank theory, the lower the fuel level in the tank, the higher is the potential of the liquid in the tank to increase its surface area exposed to air inside the tank for easy evaporation (while moving about in the tank). Because, As we know, the rate of evaporation of a liquid is directly proportional to the surface area exposed to air.
antz.bin is offline   Reply With Quote
Old 7th November 2012, 17:09   #24
Senior - BHPian
 
amalji's Avatar
 
Join Date: Sep 2009
Location: Kochi
Posts: 1,254
Thanked: 1,421 Times
Exclamation Hyundai & Kia caught cheating its customers on mileage calculation

I'm composing this post based on inputs from my co-worker - Sojish Krishnan who owns a Fluidic Verna. He is not able to post here, since he did not want to go through the process of composing an SOP for user registration.

Background

My friend decided to do a mileage test and drove the car from Trivandrum to Cochin through the MC road ( which is in top condition now and doesn't have too much traffic as well ). Most of his driving was in 6th gear with a speed of around 80 kmph. After the trip, the average mileage on the console showed as 21.05 kmpl. But, when he re-fueled, he felt that he filled more fuel than what is required. So, he did a calculation, and found the tank full to tank full mileage to be 15.15 kmpl. He repeated the same test again and the console showed the average mileage for the trip as 20 plus. This time, the actual mileage was 14.6 kmpl.

Then he searched the internet and couldn't find a reason for such huge variation. Recently, he saw an article on http://www.forbes.com/sites/jimgorze...stimating-mpg/ which says that Hyundai overestimated the mileage showed and they will reimburse the change in mileage to its owners ( only in US ). Team-BHP thread (EPA finds Hyundai & Kia overstated gas mileage ratings) on the same.

If this is true, this is as good as cheating the customers.

Question to Hyundai owners

1) Have you noticed such huge discrepancies for tank full to tank full mileage vs the average mileage shown on the console ?
2) If so, did you report it to Hyundai and what was their reaction ?

Last edited by Rehaan : 9th November 2012 at 11:46. Reason: Adding link to TBHP thread on the US issue ;)
amalji is online now   Reply With Quote
Old 7th November 2012, 17:17   #25
BHPian
 
Join Date: Jun 2012
Location: Bengaluru
Posts: 245
Thanked: 76 Times
Default Re: Hyundai & Kia caught cheating its customers on mileage calculation

The news of Company estimated MPG vs Actual MPG being fabricated by car companies to their benefits may be news in US. It's way too common in India.
All companies do it here and get away with it.

Bikes: None of the bike maker's advertised mileage comes even close to the actual.

Cars: Ford Fiesta Classic for starters advertising 32kmph.

And your mileage indicator discrepancy is strange. I havent seen such a big deviation so far. Generally its 1-3kmpl.
shadows123 is offline   Reply With Quote
Old 7th November 2012, 17:26   #26
BHPian
 
Arkesh's Avatar
 
Join Date: Sep 2012
Location: Bhubaneswar
Posts: 135
Thanked: 44 Times
Default Re: Hyundai & Kia caught cheating its customers on mileage calculation

I think the MID in Fiats is more realistic.
Never felt such discrepancy in terms of filling fuel and display of car.
I have no idea about verna though.
Arkesh is offline   Reply With Quote
Old 7th November 2012, 17:27   #27
Senior - BHPian
 
amalji's Avatar
 
Join Date: Sep 2009
Location: Kochi
Posts: 1,254
Thanked: 1,421 Times
Post Re: Hyundai & Kia caught cheating its customers on mileage calculation

Quote:
Originally Posted by shadows123 View Post
The news of Company estimated MPG vs Actual MPG being fabricated by car companies to their benefits may be news in US. It's way too common in India.
All companies do it here and get away with it.

Bikes: None of the bike maker's advertised mileage comes even close to the actual.

Cars: Ford Fiesta Classic for starters advertising 32kmph.

And your mileage indicator discrepancy is strange. I havent seen such a big deviation so far. Generally its 1-3kmpl.
1) It's not the company estimated mileage that's the issue here. ARAI mileage is done in standard test condition, and is for no load ( that is not when the car is fully loaded ). It's the ECU calculated average mileage shown on the car that is showing such huge discrepancies. This shouldn't be so if they use the correct algorithm.
2) It's not my car. It's my friend's car.
amalji is online now   Reply With Quote
Old 7th November 2012, 17:30   #28
BHPian
 
Join Date: May 2010
Location: Chandigarh
Posts: 231
Thanked: 178 Times
Default Re: Hyundai & Kia caught cheating its customers on mileage calculation

The values indicated by the mileage indicator should always be taken with a pinch of salt but the case you mention is certainly strange.
I am not sure what your friend will be able to accomplish by reporting this to Hyundai because I am quite sure that he would be given the cold shoulder.
A better alternative could be the Consumer Forum if your friend can build up a strong case.
veedub89 is offline   Reply With Quote
Old 7th November 2012, 17:31   #29
BHPian
 
arpanjha's Avatar
 
Join Date: Apr 2006
Location: Kolkata
Posts: 165
Thanked: 36 Times
Default Re: Hyundai & Kia caught cheating its customers on mileage calculation

Although a bit OT as I am not a Hyundai user, but my average FE indicator on the Punto never differs more than 1 to 1.5kpl, there might be some flaw with the sensors etc on the fluidic verna. Or maybe seriously Hyundai is taking its customers for granted.
arpanjha is offline   (1) Thanks Reply With Quote
Old 7th November 2012, 17:36   #30
Senior - BHPian
 
1100D's Avatar
 
Join Date: Jun 2006
Location: Calcutta
Posts: 4,179
Thanked: 2,358 Times
Default Re: Hyundai & Kia caught cheating its customers on mileage calculation

21kmpl shown on the console vs 15.5 actual is almost a 25% variation, too huge. Not sure if its just algorithm or something not working right.
1100D is offline   Reply With Quote
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Younger people prefer digital shimmer to accurate sound greenhorn Shifting gears 18 17th March 2010 07:34
Swine Flu Information - Important Information in post#102 on Page 7 loving_alaap Shifting gears 206 25th August 2009 14:02
How accurate are speedometers? 9216 Technical Stuff 18 21st May 2009 13:16
Whats is the accurate method of measuring FE Anju Technical Stuff 9 10th December 2007 16:52
Makes you wonder how accurate TopGear tests are(inside M6 vs 996TT) 1Day The International Automotive Scene 2 25th June 2006 12:35


All times are GMT +5.5. The time now is 07:22.

Copyright 2000 - 2017, Team-BHP.com
Proudly powered by E2E Networks